Original Article
Regional variations in hypertension prevalence and management in Germany: results from the German Health Interview and Examination Survey (DEGS1) Claudia Diederichs and Hannelore Neuhauser
Objective: This study analyzed regional differences in blood pressure (BP) distribution and management in Germany 2008–2011 in a nationwide study. Methods: The analyses were based on standardized BP measurements and anatomical therapeutic chemical classification-coded medication from the populationbased German Health Interview and Examination Survey (DEGS1) 2008–2011 (N ¼ 7074, 18–79 years, 180 study points, five regions: Central-East, South, Central-West, North-West, and North-East). Regional differences were tested between the region with the highest and lowest values. Results: Regional variations were observed in mean SBP, mean DBP, and the prevalence of hypertension in both sexes, as well as awareness, treatment, and control in men. Differences in blood pressure (in mmHg) between Central-East, the region with the highest BP level and the region with the lowest BP level, were SBP 3.2 and DBP 2.5 in men and SBP 4.5 and DBP 2.4 in women. In Central-East 39% of men and 40% of women had hypertension, versus 30% of men in the North-West and 26% of women in the South. The percentage of aware, treated, and controlled men ranged between 92, 78, and 56% in the North-East and 74, 59, and 41% in the South, respectively. After multivariate adjustment for sociodemographic variables and hypertension risk factors, geographical differences persisted for hypertension prevalence in women and hypertension awareness and treatment in men. Conclusion: So far, national surveys allowed only BP comparisons along the former East–West border and showed more elevated BP in the East. New analyses suggest regional differences with both the most and the least favorable results in the two neighboring parts of former East Germany. Keywords: awareness, control, Germany, hypertension, prevalence, regional variations, treatment Abbreviations: BP, blood pressure; DEGS1, German Health Interview and Examination Survey; ISCED, International Standard Classification of Education; KORA, Kooperative Gesundheitsforschung in der Region Augsburg Study; NHIES98, National Health Interview and Examination Survey; SHIP, Study of Health in Pomerania
INTRODUCTION
H
ypertension is one of the most important risk factors for cardiovascular diseases and resulting morbidity and mortality [1]. However, despite the availability of efficient therapies and the dramatic improvement of hypertension treatment during the last 30 years [2], an effective prevention and control of high blood pressure (BP), which is essential in the prevention of myocardial infarction, stroke, and heart and renal failure [3], remains largely unsatisfactory [4]. Differences in hypertension prevalence by sex, age, race, and sociodemographic characteristics are well researched [5–7]. Furthermore, strong geographical variations of BP levels have been reported between countries [8] and between regions in Austria [9], Germany [10], the United States [7], China [11], France [12], and Croatia [13]. Considerable geographical variations at the country level were also observed for hypertension awareness, treatment, and control rates [14]. Within Germany, data from two population-based national surveys conducted by the Robert Koch Institute in 1991 and 1998 suggested that mean SBP and mean DBP were higher in the former East compared with the West [15,16]. Furthermore, a predominantly rural region in Mecklenburg-Vorpommern in the North-East was found to have a much higher mean SBP and DBP than the region around the city of Augsburg in Bavaria in the South-West. This was shown in a comparison of data from the Study of Health in Pomerania (SHIP) 1997–2000 and the Kooperative Gesundheitsforschung in der Region Augsburg Study (KORA) 2000, which used the same BP measurement device and protocol. Interestingly, no geographical variations in the percentage of aware, treated, and controlled people with high BP were observed between the two regions [10]. Journal of Hypertension 2014, 32:1405–1414 Robert Koch Institute and German Centre for Cardiovascular Research (DZHK), Partner Site, Berlin, Germany Correspondence to Claudia Diederichs, Robert Koch Institute, General-Pape-Strasse 62-66, 12101 Berlin, Germany. Tel: +49 30 18754 3218; fax: +49 30 18754 3555; e-mail: [email protected] Received 19 December 2013 Revised 17 March 2014 Accepted 18 March 2014 J Hypertens 32:1405–1414 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. DOI:10.1097/HJH.0000000000000211
Journal of Hypertension
www.jhypertension.com
1405
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diederichs and Neuhauser
More than 20 years after the German reunification, the most recent mortality data show persisting regional differences in both ischemic heart disease and stroke mortality [17]. In 2010, the highest age-standardized death rates for cerebrovascular diseases were found in Sachsen and Sachsen-Anhalt for men and Sachsen-Anhalt and Saarland for women. Noticeably low rates for both sexes were reported in the north, namely Schleswig-Holstein and Hamburg. However, recent data on the regional distribution of BP, which is a major determinant of cardiovascular and, in particular, of cerebrovascular mortality were lacking. Therefore, we used data from the 2008–2011 populationbased national German Health Interview and Examination Survey (DEGS1) to analyze regional differences in mean BP level, hypertension prevalence, awareness, treatment, and control within Germany.
METHODS Study design and study population The German Health Interview and Examination Survey 2008–2011 (DEGS1) is a national, population-based health survey of adults between 18 and 79 years living in Germany. Detailed information on the study population, sampling, and measurement procedures has been described in a previous publication [18]. Briefly, the overall objective of DEGS1 is to analyze the health status, health risks and resources, functional capacity levels, and disability in the German population. It combines longitudinal data by recruiting former participants of the 1998 National Health Interview and Examination Survey (NHIES98) and a crosssectional design with newly sampled individuals. A two-stage stratified clustered sampling plan was used to select 120 communities (in NHIES98) and 60 additional communities (in DEGS1) as primary sampling units from a list of German communities stratified according to federal states and the type of community [19]. Within selected communities, random samples of individuals, stratified by 10-year age group, were drawn from local population registers. Sample weights were applied to adjust different sampling probabilities within the strata and correct deviations between the design-weighted net sample and German standard population from 31 December 2010, with respect to age, sex, level of education, status as a foreigner, federal state, and type of community [18]. Between November 2008 and November 2011, 3959 former NHIES89 participants (response rate 62%) and 4193 newly sampled persons (response rate 42%) took part in the survey, which included several self-administered questionnaires, a standardized medical interview, the collection of urine and blood samples, as well as standardized measurements and physical performance tests. For this analysis, we used data from 7074 individuals, who were between 18 and 79 years old and who had complete BP and medication data. All of the participants gave informed consent.
Collection of information Baseline data on sociodemographic variables, such as age, highest school qualification [20], physical activity [21], alcohol consumption [22], and the utilization of healthcare 1406
www.jhypertension.com
services [23] were collected in a self-administered questionnaire. During a standardized, computer-assisted medical interview, participants were asked whether high or elevated BP had ever been diagnosed by a physician. Medication use including indication, dose, frequency, and duration of use was documented on the basis of original drug container brought to the survey site and coded according to the WHO anatomical therapeutic chemical classification system. As part of the physical examination, body weight and body height were measured in underwear with shoes removed according to a standardized protocol and using portable electronic scales (Sa¨ulenwaage 930; SECA, Hamburg, Germany) and stadiometer (Harpenden Portable Stadiometer; Holtain Ltd, Crosswell, Crymych, Pembrokeshire, UK) [24]. Three consecutive automated BP measurements (Datascope Accutor Plus, Mindray DS; Redmond, Washington, USA) were taken at 3 min interval, after the participants sat and relaxed for at least 5 min on a height adjustable chair with their back supported. The elbow was slightly bent and lying on a table at the level of the right atrium. Both feet were straight on the floor and legs were not crossed. Three different cuff sizes were used: bladder size 10.6 23.9 cm for upper arm circumferences 21.0–27.9 cm; 13.5 30.7 cm for arm circumferences 28.0–35.9 cm; and bladder size 17.0 38.6 cm for arm circumferences 36.0– 46.0 cm. The second and third measurements were averaged [25].
Definitions Hypertension was defined as a mean SBP 140 mmHg at least, or a mean DBP 90 mmHg at least, or the use of antihypertensive medication, given that the participants were aware of having hypertension. Hypertensive participants were considered aware of their condition if they reported that they had ever been told by a doctor that they have high or elevated BP. Hypertensive individuals were identified as treated if they currently used antihypertensive medication [antihypertensives (anatomical therapeutic chemical code C02), diuretics (C03), b-blocking agents (C07), calcium channel blockers (C08), or agents acting on the renin–angiotensin system (C09)] and were aware of hypertension, and they were classified as controlled if they had an average SBP lower than 140 mmHg and an average DBP lower than 90 mmHg. BMI was defined as measured weight in kilograms divided by squared height in meters.
Regional classification The 16 states in Germany were grouped into five regions considering the former inner-German border (Fig. 1 and by assigning three federal city states Berlin, Bremen, and Hamburg consistently to their surrounding neighboring states. 1. North-West (Schleswig-Holstein, Hamburg, Niedersachsen, Bremen) 2. North-East (Mecklenburg-Vorpommern, Brandenburg, Berlin) 3. Central-West (Nordrhein-Westfalen, Hessen, Rheinland-Pfalz, Saarland) 4. Central-East (Sachsen-Anhalt, Sachsen, Thu¨ringen) 5. South (Bayern, Baden-Wu¨rttemberg) Volume 32 Number 7 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Hypertension in Germany
A comparison of the five regional DEGS1-subsamples with national statistics on 31 December 2010 [26,27] with regard to age, sex, level of education, and community type is given in Table 1. According to the ‘BIK classification system (BIK ASCHPURWIS þ BEHRENS GmbH, Hamburg, Germany), 10 different types of communities were identified on – in simple terms – the basis of their population density and commuter rate. The BIK-classes were grouped into BIK 1–7 and BIK 8–10 [19]. The level of education was classified into low [International Standard Classification of Education (ISCED): 1, 2], medium (ISCED: 3B, 3A), or high (ISCED: 4B, 4A, 5A) according to the ISCED [28].
Data analysis Sex-specific means of SBP and DBP and the prevalence of hypertension, as well as the percentages of aware, treated, and controlled hypertensive individuals were stratified by region. To take the different age distribution of the federal states into account, we compared the results with BP means and hypertension prevalence directly standardized to the German population from 31 December 2010 (results not shown in detail but only summarized in the text). Differences between categorical variables were estimated with the x2 test and between continuous variables with the t test. Statistical significance of regional differences was tested among the region with the highest and lowest point estimates (Table 2) at the P 0.05 level. The region with the lowest prevalence of hypertension and the lowest management performance was defined as reference group in the multivariate analysis (Tables 3–6). We adjusted the models based on known risk factors, including age (18–39, 40–59,
FIGURE 1 Five regions in Germany.
TABLE 1. Comparison between national population statistics and weighted German Health Interview and Examination Survey sample with respect to age, education, and community type (in %) Central-East Region Men Age
Education levela
Women Age
Education levela
Both sexes Community typeb
South
Central-West
North-West
North-East
Population DEGS1 Population DEGS1 Population DEGS1 Population DEGS1 Population DEGS1 18–29 30–39 40–49 50–59 60–69 70–79 Low Medium High
18.2 14.8 20.0 19.6 14.5 12.7 9.8 60.6 29.6
18.4 14.6 20.7 19.6 14.6 12.7 8.3 62.9 28.9
19.1 16.0 22.4 17.9 13.4 11.0 16.2 50.2 33.6
18.2 17.0 22.4 17.9 13.4 11.0 17.7 48.1 34.2
18.6 15.3 22.1 18.5 13.6 11.5 19.3 49.1 31.6
19.1 14.8 22.2 18.6 13.7 11.6 20.4 49.4 30.2
18.3 15.7 22.3 17.5 14.2 11.7 18.8 51.8 29.4
21.7 12.5 22.5 17.2 14.3 11.8 16.5 47.4 36.1
18.6 15.9 21.6 18.9 13.5 11.2 14.5 51.4 34.1
19.9 14.6 21.6 19.0 13.5 11.3 10.4 57.2 32.4
18–29 30–39 40–49 50–59 60–69 70–79 Low Medium High
16.3 12.9 18.8 19.5 15.9 16.3 15.5 56.3 28.3
16.6 12.6 18.9 19.6 15.9 16.4 14.2 59.1 26.7
18.3 15.7 21.2 17.6 13.8 13.0 30.5 47.7 21.8
19.2 14.9 21.3 17.7 13.9 13.1 27.5 51.0 21.5
17.7 15.0 21.0 18.2 14.1 13.7 31.1 46.5 22.4
17.5 15.2 21.1 18.2 14.2 13.8 27.9 46.5 25.6
17.6 15.2 21.1 17.4 14.6 13.8 29.7 48.8 21.5
19.2 13.6 21.2 17.4 14.7 13.9 28.1 47.2 24.7
17.7 14.7 20.2 18.6 14.4 14.1 19.5 49.1 31.4
19.8 12.6 20.3 18.7 14.5 14.2 20.3 46.3 33.4
BIK 1–7 BIK 8–10
65.5 34.5
65.5 34.5
55.5 44.5
55.5 44.5
46.5 53.4
46.6 53.4
46.9 53.1
47.0 52.0
39.0 61.0
39.0 61.0
DEGS1, German Health Interview and Examination Survey; ISCED, International Standard Classification of Education. a ISCED-education level (low, ISCED 1; 2 versus medium, ISCED 3B; 3A versus high, ISCED 4B, 4A, 5A). b BIK-classification (1,
Regional variations in hypertension prevalence and management in Germany: results from the German Health Interview and Examination Survey (DEGS1) Claudia Diederichs and Hannelore Neuhauser
Objective: This study analyzed regional differences in blood pressure (BP) distribution and management in Germany 2008–2011 in a nationwide study. Methods: The analyses were based on standardized BP measurements and anatomical therapeutic chemical classification-coded medication from the populationbased German Health Interview and Examination Survey (DEGS1) 2008–2011 (N ¼ 7074, 18–79 years, 180 study points, five regions: Central-East, South, Central-West, North-West, and North-East). Regional differences were tested between the region with the highest and lowest values. Results: Regional variations were observed in mean SBP, mean DBP, and the prevalence of hypertension in both sexes, as well as awareness, treatment, and control in men. Differences in blood pressure (in mmHg) between Central-East, the region with the highest BP level and the region with the lowest BP level, were SBP 3.2 and DBP 2.5 in men and SBP 4.5 and DBP 2.4 in women. In Central-East 39% of men and 40% of women had hypertension, versus 30% of men in the North-West and 26% of women in the South. The percentage of aware, treated, and controlled men ranged between 92, 78, and 56% in the North-East and 74, 59, and 41% in the South, respectively. After multivariate adjustment for sociodemographic variables and hypertension risk factors, geographical differences persisted for hypertension prevalence in women and hypertension awareness and treatment in men. Conclusion: So far, national surveys allowed only BP comparisons along the former East–West border and showed more elevated BP in the East. New analyses suggest regional differences with both the most and the least favorable results in the two neighboring parts of former East Germany. Keywords: awareness, control, Germany, hypertension, prevalence, regional variations, treatment Abbreviations: BP, blood pressure; DEGS1, German Health Interview and Examination Survey; ISCED, International Standard Classification of Education; KORA, Kooperative Gesundheitsforschung in der Region Augsburg Study; NHIES98, National Health Interview and Examination Survey; SHIP, Study of Health in Pomerania
INTRODUCTION
H
ypertension is one of the most important risk factors for cardiovascular diseases and resulting morbidity and mortality [1]. However, despite the availability of efficient therapies and the dramatic improvement of hypertension treatment during the last 30 years [2], an effective prevention and control of high blood pressure (BP), which is essential in the prevention of myocardial infarction, stroke, and heart and renal failure [3], remains largely unsatisfactory [4]. Differences in hypertension prevalence by sex, age, race, and sociodemographic characteristics are well researched [5–7]. Furthermore, strong geographical variations of BP levels have been reported between countries [8] and between regions in Austria [9], Germany [10], the United States [7], China [11], France [12], and Croatia [13]. Considerable geographical variations at the country level were also observed for hypertension awareness, treatment, and control rates [14]. Within Germany, data from two population-based national surveys conducted by the Robert Koch Institute in 1991 and 1998 suggested that mean SBP and mean DBP were higher in the former East compared with the West [15,16]. Furthermore, a predominantly rural region in Mecklenburg-Vorpommern in the North-East was found to have a much higher mean SBP and DBP than the region around the city of Augsburg in Bavaria in the South-West. This was shown in a comparison of data from the Study of Health in Pomerania (SHIP) 1997–2000 and the Kooperative Gesundheitsforschung in der Region Augsburg Study (KORA) 2000, which used the same BP measurement device and protocol. Interestingly, no geographical variations in the percentage of aware, treated, and controlled people with high BP were observed between the two regions [10]. Journal of Hypertension 2014, 32:1405–1414 Robert Koch Institute and German Centre for Cardiovascular Research (DZHK), Partner Site, Berlin, Germany Correspondence to Claudia Diederichs, Robert Koch Institute, General-Pape-Strasse 62-66, 12101 Berlin, Germany. Tel: +49 30 18754 3218; fax: +49 30 18754 3555; e-mail: [email protected] Received 19 December 2013 Revised 17 March 2014 Accepted 18 March 2014 J Hypertens 32:1405–1414 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. DOI:10.1097/HJH.0000000000000211
Journal of Hypertension
www.jhypertension.com
1405
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diederichs and Neuhauser
More than 20 years after the German reunification, the most recent mortality data show persisting regional differences in both ischemic heart disease and stroke mortality [17]. In 2010, the highest age-standardized death rates for cerebrovascular diseases were found in Sachsen and Sachsen-Anhalt for men and Sachsen-Anhalt and Saarland for women. Noticeably low rates for both sexes were reported in the north, namely Schleswig-Holstein and Hamburg. However, recent data on the regional distribution of BP, which is a major determinant of cardiovascular and, in particular, of cerebrovascular mortality were lacking. Therefore, we used data from the 2008–2011 populationbased national German Health Interview and Examination Survey (DEGS1) to analyze regional differences in mean BP level, hypertension prevalence, awareness, treatment, and control within Germany.
METHODS Study design and study population The German Health Interview and Examination Survey 2008–2011 (DEGS1) is a national, population-based health survey of adults between 18 and 79 years living in Germany. Detailed information on the study population, sampling, and measurement procedures has been described in a previous publication [18]. Briefly, the overall objective of DEGS1 is to analyze the health status, health risks and resources, functional capacity levels, and disability in the German population. It combines longitudinal data by recruiting former participants of the 1998 National Health Interview and Examination Survey (NHIES98) and a crosssectional design with newly sampled individuals. A two-stage stratified clustered sampling plan was used to select 120 communities (in NHIES98) and 60 additional communities (in DEGS1) as primary sampling units from a list of German communities stratified according to federal states and the type of community [19]. Within selected communities, random samples of individuals, stratified by 10-year age group, were drawn from local population registers. Sample weights were applied to adjust different sampling probabilities within the strata and correct deviations between the design-weighted net sample and German standard population from 31 December 2010, with respect to age, sex, level of education, status as a foreigner, federal state, and type of community [18]. Between November 2008 and November 2011, 3959 former NHIES89 participants (response rate 62%) and 4193 newly sampled persons (response rate 42%) took part in the survey, which included several self-administered questionnaires, a standardized medical interview, the collection of urine and blood samples, as well as standardized measurements and physical performance tests. For this analysis, we used data from 7074 individuals, who were between 18 and 79 years old and who had complete BP and medication data. All of the participants gave informed consent.
Collection of information Baseline data on sociodemographic variables, such as age, highest school qualification [20], physical activity [21], alcohol consumption [22], and the utilization of healthcare 1406
www.jhypertension.com
services [23] were collected in a self-administered questionnaire. During a standardized, computer-assisted medical interview, participants were asked whether high or elevated BP had ever been diagnosed by a physician. Medication use including indication, dose, frequency, and duration of use was documented on the basis of original drug container brought to the survey site and coded according to the WHO anatomical therapeutic chemical classification system. As part of the physical examination, body weight and body height were measured in underwear with shoes removed according to a standardized protocol and using portable electronic scales (Sa¨ulenwaage 930; SECA, Hamburg, Germany) and stadiometer (Harpenden Portable Stadiometer; Holtain Ltd, Crosswell, Crymych, Pembrokeshire, UK) [24]. Three consecutive automated BP measurements (Datascope Accutor Plus, Mindray DS; Redmond, Washington, USA) were taken at 3 min interval, after the participants sat and relaxed for at least 5 min on a height adjustable chair with their back supported. The elbow was slightly bent and lying on a table at the level of the right atrium. Both feet were straight on the floor and legs were not crossed. Three different cuff sizes were used: bladder size 10.6 23.9 cm for upper arm circumferences 21.0–27.9 cm; 13.5 30.7 cm for arm circumferences 28.0–35.9 cm; and bladder size 17.0 38.6 cm for arm circumferences 36.0– 46.0 cm. The second and third measurements were averaged [25].
Definitions Hypertension was defined as a mean SBP 140 mmHg at least, or a mean DBP 90 mmHg at least, or the use of antihypertensive medication, given that the participants were aware of having hypertension. Hypertensive participants were considered aware of their condition if they reported that they had ever been told by a doctor that they have high or elevated BP. Hypertensive individuals were identified as treated if they currently used antihypertensive medication [antihypertensives (anatomical therapeutic chemical code C02), diuretics (C03), b-blocking agents (C07), calcium channel blockers (C08), or agents acting on the renin–angiotensin system (C09)] and were aware of hypertension, and they were classified as controlled if they had an average SBP lower than 140 mmHg and an average DBP lower than 90 mmHg. BMI was defined as measured weight in kilograms divided by squared height in meters.
Regional classification The 16 states in Germany were grouped into five regions considering the former inner-German border (Fig. 1 and by assigning three federal city states Berlin, Bremen, and Hamburg consistently to their surrounding neighboring states. 1. North-West (Schleswig-Holstein, Hamburg, Niedersachsen, Bremen) 2. North-East (Mecklenburg-Vorpommern, Brandenburg, Berlin) 3. Central-West (Nordrhein-Westfalen, Hessen, Rheinland-Pfalz, Saarland) 4. Central-East (Sachsen-Anhalt, Sachsen, Thu¨ringen) 5. South (Bayern, Baden-Wu¨rttemberg) Volume 32 Number 7 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Hypertension in Germany
A comparison of the five regional DEGS1-subsamples with national statistics on 31 December 2010 [26,27] with regard to age, sex, level of education, and community type is given in Table 1. According to the ‘BIK classification system (BIK ASCHPURWIS þ BEHRENS GmbH, Hamburg, Germany), 10 different types of communities were identified on – in simple terms – the basis of their population density and commuter rate. The BIK-classes were grouped into BIK 1–7 and BIK 8–10 [19]. The level of education was classified into low [International Standard Classification of Education (ISCED): 1, 2], medium (ISCED: 3B, 3A), or high (ISCED: 4B, 4A, 5A) according to the ISCED [28].
Data analysis Sex-specific means of SBP and DBP and the prevalence of hypertension, as well as the percentages of aware, treated, and controlled hypertensive individuals were stratified by region. To take the different age distribution of the federal states into account, we compared the results with BP means and hypertension prevalence directly standardized to the German population from 31 December 2010 (results not shown in detail but only summarized in the text). Differences between categorical variables were estimated with the x2 test and between continuous variables with the t test. Statistical significance of regional differences was tested among the region with the highest and lowest point estimates (Table 2) at the P 0.05 level. The region with the lowest prevalence of hypertension and the lowest management performance was defined as reference group in the multivariate analysis (Tables 3–6). We adjusted the models based on known risk factors, including age (18–39, 40–59,
FIGURE 1 Five regions in Germany.
TABLE 1. Comparison between national population statistics and weighted German Health Interview and Examination Survey sample with respect to age, education, and community type (in %) Central-East Region Men Age
Education levela
Women Age
Education levela
Both sexes Community typeb
South
Central-West
North-West
North-East
Population DEGS1 Population DEGS1 Population DEGS1 Population DEGS1 Population DEGS1 18–29 30–39 40–49 50–59 60–69 70–79 Low Medium High
18.2 14.8 20.0 19.6 14.5 12.7 9.8 60.6 29.6
18.4 14.6 20.7 19.6 14.6 12.7 8.3 62.9 28.9
19.1 16.0 22.4 17.9 13.4 11.0 16.2 50.2 33.6
18.2 17.0 22.4 17.9 13.4 11.0 17.7 48.1 34.2
18.6 15.3 22.1 18.5 13.6 11.5 19.3 49.1 31.6
19.1 14.8 22.2 18.6 13.7 11.6 20.4 49.4 30.2
18.3 15.7 22.3 17.5 14.2 11.7 18.8 51.8 29.4
21.7 12.5 22.5 17.2 14.3 11.8 16.5 47.4 36.1
18.6 15.9 21.6 18.9 13.5 11.2 14.5 51.4 34.1
19.9 14.6 21.6 19.0 13.5 11.3 10.4 57.2 32.4
18–29 30–39 40–49 50–59 60–69 70–79 Low Medium High
16.3 12.9 18.8 19.5 15.9 16.3 15.5 56.3 28.3
16.6 12.6 18.9 19.6 15.9 16.4 14.2 59.1 26.7
18.3 15.7 21.2 17.6 13.8 13.0 30.5 47.7 21.8
19.2 14.9 21.3 17.7 13.9 13.1 27.5 51.0 21.5
17.7 15.0 21.0 18.2 14.1 13.7 31.1 46.5 22.4
17.5 15.2 21.1 18.2 14.2 13.8 27.9 46.5 25.6
17.6 15.2 21.1 17.4 14.6 13.8 29.7 48.8 21.5
19.2 13.6 21.2 17.4 14.7 13.9 28.1 47.2 24.7
17.7 14.7 20.2 18.6 14.4 14.1 19.5 49.1 31.4
19.8 12.6 20.3 18.7 14.5 14.2 20.3 46.3 33.4
BIK 1–7 BIK 8–10
65.5 34.5
65.5 34.5
55.5 44.5
55.5 44.5
46.5 53.4
46.6 53.4
46.9 53.1
47.0 52.0
39.0 61.0
39.0 61.0
DEGS1, German Health Interview and Examination Survey; ISCED, International Standard Classification of Education. a ISCED-education level (low, ISCED 1; 2 versus medium, ISCED 3B; 3A versus high, ISCED 4B, 4A, 5A). b BIK-classification (1,
