Diabetes Prevention and Care

The EUCID project carried out pilot diabetes data collection

The aim of the European Core Indicators in Diabetes (EUCID) project was to collect and compare data on risk factors for diabetes, complications and quality of care indicators in (future) Member States of the European Union.

19 countries provided data for a list of 35 indicators. The data were representative at a regional or a national level, for 2004, 2005 or 2006, and were specified by age band. The precise data collection methods (surveys, registries, administrative databases) were provided. Data were age-standardized to allow comparison with the general population where possible, or with the representative regional population if the national level was not possible.

The results show large variation in coverage between the indicators

For some of the 35 indicators data was available from most of the participating countries, for others data was only available from a small number of countries. Some striking results are summarized below:

• Among the least available indicators was the incidence of blindness in people with diabetes, which was provided by only 4 countries, and impaired fasting glucose in the general population, which was provided by only 2.
• The standardized prevalence of diabetes varied from 2.6% in Finland to 7.6% in Cyprus; crude incidence of diabetes (0-14 yrs) from 11 per 100,000 in Spain to 60 per 100,000 in Finland; standardized prevalence of overweight (25-74 yrs) from 37% in Germany to 60% in Cyprus; standardized mortality rates linked with diabetes from 7 per 100,000 in Luxembourg to 56 per 100,000 in Finland.
• Among people with diabetes (>25 yrs), process indicators ranged: for HbA1c testing once a year, from 51% in Ireland to 99% in the Netherlands, France and Belgium; for lipid testing, from 45% in Ireland to 99% in the Netherlands; for microalbuminuria testing, from 25% in Finland to 97% in the Netherlands; for eye examination, from 12% in Ireland to 84% in the Netherlands.
• Risk factors in people with diabetes varied also. HbA1c>7%: 32% in Ireland to 83% in Cyprus; total cholesterol>5mmol/l: 14% in Ireland to 68% in Cyprus; microalbuminuria (enhanced protein levels in the urine): 9% in Finland to 41% in England; blood pressure>140/90mmHg: 17% in France to 46% in Sweden; smoking: 10% in Ireland to 37% in Denmark.
• Complication incidence rates were: dialysis and transplantation: 4 per 100,000 diabetes clients in Cyprus to 149 in Scotland; stroke: 37 in Cyprus to 2675 in Germany; myocardial infarction: 21 in Cyprus to 2135 in Austria; major amputation: 78 in Scotland to 574 in Spain.

Substantial improvements of comparable data collection possible

While many European countries can provide data for the list of diabetes indicators, there are gaps for some important items such as diabetes-linked blindness. Most of the European countries achieve remarkable good testing of people with diabetes. Risk factors and outcomes vary across countries, reflecting a mixture of genetic background, societal and cultural factors, as well as public health policies. Although the data collection has been carried out quite carefully, some of the large differences found may reflect methodological variations. The results of EUCID will be used within the countries both to try to influence policies with respect to interventions and to improve data collection.

Separate EUphacts are presented on Diabetes, Overweight, Smoking and Blood Pressure

This EUphocus presents data related to overweight, smoking and blood pressure, supplementing the more general data provided in the EUphacts Diabetes, Overweight, Smoking and Blood pressure. The data shown in those EUphacts have been obtained from regular sources, and sometimes differ from the data shown in this EUphocus that have been collected by EUCID.

Diabetes is a large and increasing disease burden in Europe and beyond

Diabetes mellitus is a growing burden for all the countries in the world. The International Diabetes Federation (IDF) estimates that the number of people with diabetes will grow from 194 million in 2003 to 333 million in 2025 (IDF, 2007). The increase is expected to particularly occur in developing countries, but also in industrialized countries. The estimates for 2025 were based on projections of observed increases in prevalence.

Not only the diagnosis of diabetes and the treatment of its characteristic elevated blood glucose (hyperglycemia) are important for reducing the societal burden of diabetes, but also the diagnosis and treatment of complications that may follow from diabetes. The most important of these are microvascular disease (neuropathy, retinopathy and nephropathy) and macrovascular disease (heart, cerebral and peripheral vessels). The actual numbers for these complications are not very well known at the national level, restricting the validity of international comparisons.

A growing proportion of all populations in the world present a state of impaired glucose tolerance or impaired fasting glucose, and are therefore at risk of developing diabetes. About 10% of this group will develop diabetes annually. The IDF estimates this number to be 50% higher than the population with diabetes. For more detailed information see the EUpact on Diabetes.

The EUCID project is a follow-up of the EUDIP project

The European Diabetes Indicators Project (EUDIP) in 2000-2003 defined and piloted diabetes risk and diabetes care indicators on the national level in EU Member states, and as a result published a list of feasible indicators. The European Core Indicators in Diabetes (EUCID) project collected these indicators for the year 2005 (+/- 1 year) from 19 countries. Both projects were co-financed by DG-SANCO under the Health Monitoring Programme and the Public Health Programme, respectively.

The list of 35 indicators can be seen in Overview of data availability per indicator by country. The entire EUCID report, including annexes, can be viewed and downloaded at www.eucid.eu.

EUCID participants come from 19 countries

Efforts were made to include all (future) EU Member States in the project, via diabetes associations, diabetes health care professionals and governments. This resulted in the actual participation of the 19 partners shown in the List of partners collaborating in EUCID. This includes 18 EU Member States and Turkey.

The collected data come from a variety of sources

For most countries or indicators, complete national data were not available, but often data could be provided from a regional source. The data sources were divided into four groups: (1) national complete, (2) national sample, (3) regional complete, (4) regional sample. For the indicators concerning the diabetic population, five different sources could be discerned: (a) administrative databases, (b) national surveys, (c) clinical database, primary care, (d) clinical database, primary + secondary care, and (e) clinical database, secondary care. All relevant properties of the national and regional data sources are listed in the code book (annex one of the EUCID report).

In order to allow age-standardization, data were collected, where possible by 10-year age bands. Standardization was done according to the European Standard Population (Waterhouse et al., 1976), when national data for the general population were available. Diabetes population data could not be standardized since there is no standard population for diabetes .

Data availability showed large variations between indicators and between countries

It appeared that the availability of data varied considerably between indicators and between countries, as shown in Overview of data availability per country. For the in depth analysis per country see the EUCID report.

From the table, it can be seen that some indicators were available for almost all countries, such as prevalence of diabetes, while others were almost completely unavailable, such as timely laser treatment for diabetic retinopathy.

Data collection has been accurate, but improvements are possible

For some indicators, the data sources were different in different countries, so that the comparability of the indicators is not optimal. In the area of testing people with diabetes, some data originated from databases that might not reflect the average situation.

The future EUBIROD (European Best Indicators through Regional Outcomes Diabetes) project will combine national and regional indicators in an automated way, so that care planners can always have reliable indicators at their disposal. In this project two types of indicator data will be collected:

• national data on risk factors and prevalence and incidence of diabetes and major complications like stroke, blindness and kidney function replacement therapy;
• data on regional or even local quality and quantity of care from clinical databases like indicators on blood pressure and average blood glucose.

Examples from EUCID data: diabetes prevalence and incidence, overweight and diabetes mortality

This section focuses on data collected in the general population. In total, five indicators are included in this category:

• Prevalence of diabetes
• Incidence of diabetes age 0-14 years
• BMI (body mass index)
• Impaired fasting glucose
• Mortality due to diabetes

As examples, data are reproduced here on the prevalence and incidence of diabetes, on overweight, and on diabetes mortality.

Prevalence of diabetes ranges 3.3% - 7.3%

The population prevalence of diabetes ranges from 3.3% to 7.3% and from 2.6% to 7.6% when age-standardized. Three different data sources were used. See Prevalence of diagnosed diabetes in the general population, all ages per 1000 individuals and Prevalence of diagnosed diabetes in the general population, all ages, standardized, per 1000 individuals.

The prevalence rates shown in the EUphact on Diabetes, derived from the estimates of the IDF, range roughly between 5% and 10%. This is substantially higher than the values found by EUCID. The reason is that the IDF data are based on the oral glucose tolerance test and thus include previously undiagnosed diabetics.

Annual incidence in ages 0-14 ranges from 15 to 60 per 100,000 of the population

For the annual incidence ages 0-14, see ‘Crude annual incidence rate of diabetes /100.000 population 0-14 years and Standardised annual incidence rate of diabetes /100.000 population 0-14 years.

12 Countries were able to provide data on the incidence of diabetes in children 0 to 14 years of age in the EUCID project. There is a wide variation in the incidence of diabetes in children reported amongst the countries. After standardisation, the countries that are historically known for high incidence are Finland and Sweden, but Scotland and Romania as having a high incidence came as a surprise. When subdivided, the data show an increase in incidence with age until 10-14 years old. Scotland had the highest incidence in the group of 10-14 years of age.

Only 5 countries were able to provide data specifically for type 2 diabetes in children. The increase in type 2 diabetes in children is reported to be a growing problem.

Prevalence of overweight, including obesity, ranges 38% - 59%

For the data on overweight and obesity, see Crude prevalence of overweight and obesity in the general population aged 25-74 years and Standardized prevalence of overweight and obesity in the general population aged 25-74 years.

9 Countries were able to provide data for the EUCID project that could be used for the comparison. Data are shown for Body Mass Index (BMI), defined as body weight divided by the square of body length (kg/m2). On average almost 40% of the population has a BMI>25.0. Notably for overweight (BMI 25.0-29.9), the prevalence was higher for men than for women. This health problem is one of the most important risk factors for the development of type 2 diabetes. It is the only risk factor for diabetes that can be influenced, although this is very difficult. Most of the data provided was taken from Health Interview Surveys, which often lead to an underestimation of obesity and overweight. Only two countries provided data that originated from Health Examination Surveys, which do not appear to be outliers.

The prevalence rates shown in the EUphact on Overweight are derived from the IOTF. There is not too much consistency between the two data sets, but it should be noted that in the IOTF database the age ranges are not standard, and that for some countries the sources are different (self-reported versus measured). Notably the IOTF overweight figures for England, Germany and Luxemburg are quite a lot higher than the EUCID figures for these countries. For Germany and Luxemburg, this may be due to the IOTF data being based on actual measurement, as opposed to the self-report data collected by EUCID. Both databases do, however, show that overweight is more frequent in men than in women.

Mortality from diabetes is difficult to assess

Standard mortality statistics are based on the registered primary cause of death. For diabetes this gives a severe underestimation of the impact of the disease on mortality, since in many deaths from other causes, such as cardiovascular disease, diabetes plays a role and is registered as the ‘secondary cause of death’. Therefore, EUCID has collected data on mortality for diabetes as a primary and/or secondary cause of death.

Ten countries were able to provide data on this indicator for the EUCID project, as shown in Standardized Mortality Rates for Diabetes. For comparison, the table shows the Standardized Death Rates (SDR) for diabetes as a primary cause of death, from Eurostat.

The table shows that in most countries the addition of secondary causes of death leads to a (up to eight times) higher SDR, but that the impact of this addition varies per country. This variance is due to inconsistent ways of recording secondary causes of death. Diabetes may, for example, sometimes not be recorded as a secondary cause of death, due to a lack of awareness. The different levels of reimbursement offered for hospital expenses if a second diagnosis is present can also influence recording habits. The only way to obtain reliable data for this indicator would be to connect a national diabetes database with the national death register. The DARTS project in Scotland has the possibility to do this, as do some Scandinavian countries and France.

Examples from EUCID data: HbA1c and cholesterol levels in diabetics

This section focuses on data collected on biomedical risk facors in diabetics. In total, twelve indicators are included in this EUCID category:

• HbA1c tested
• HbA1c > 7.0 %
• Cholesterol tested
• Cholesterol > 5 mmol/l
• LDL-Cholesterol tested
• LDL-Cholesterol > 2.6 mmol/l
• HDL-Cholesterol tested
• HDL-Cholesterol <1.0 men and <1.25 women
• Triglycerides tested
• Triglycerides < 2.3 mmol/l
• Albuminuria tested
• Albuminuria abnormal

All of these biomedical factors have a well defined normal value and a cut off point where intervention is agreed upon. The consensus is that all of these risk factors should be measured at least annually. Most of the data originated from clinical databases, as national or regional samples that were more or less representative for the national population. The complete set of indicators can be found in the final report of the EUCID project.

Most of the European countries achieve remarkably good testing levels of people with diabetes. However some data originates from databases that might not reflect the average situation in the countries.

In general, the variation of risk factor and outcome levels across countries reflect a mixture of genetic background, societal and cultural factors, as well as public health policies, in combination with the quality and organization of local health care.

As examples for EUPHIX, data are reproduced on the proportion of the diabetics tested that showed elevated levels of HbA1c (>7.0%) or total cholesterol (> 5 mmol/l).

Among diabetics tested the percentage with HbA1c > 7% ranges from 32% to 83%

It is important for diabetes patients to reach an optimal blood glucose level, both for their general well being and to minimise the risk of micro-vascular complications. In non-diabetics, the protein HbA1c ranges from 4 to 6%. Its level reflects the average blood glucose during the last 8 to 12 weeks and is therefore a good marker for the appropriate regulation of blood glucose . Most guidelines recommend that diabetics maintain an HbA1c level below 7%, although some even recommend a threshold as low as 6.5%.

A related set of data from EUCID shows that between 51% (Ireland) and 99% (several countries) of diabetics are regularly tested for their HbA1c levels.

For the data on the present indicator, see HbA1c>7.0% in diabetic population aged >25. 11 Countries were able to provide these data for the EUCID project. The percentage of the diabetic population that did not achieve the 7% goal differs considerably between countries: in Ireland 32% did not achieve this goal versus 83% in Cyprus.

Among diabetics tested the percentage with a total cholesterol of >5 mmol/l ranged between 14% and 68%

There is much discussion about the best way to specify the risk of enhanced cholesterol levels for macro-vascular disease. A total cholesterol level above 5.0 mmol/l is, however, acknowledged by many guidelines as a screening target. The majority of diabetics who receive pharmacological treatment can reach this target.

A related set of data from EUCID shows that between 45% (Ireland) and 99% (the Netherlands) of diabetics are regularly tested for their total cholesterol levels.

For the data on the present indicator, see Cholesterol > 5 mmol/l in diabetic population aged >25. 12 Countries were able to provide these data for the EUCID project. The percentage of the diabetic population that did not achieve the 5 mmol/l goal differs considerably between countries: in Ireland and Sweden 14% did not achieve this goal versus 68% in Cyprus.

Examples from EUCID data: blood pressure, overweight and smoking in diabetics

This section focuses on data collected on physical and behavioural risk factors in diabetics. In total, six indicators are included in this EUCID category:

• Blood pressure tested
• Blood pressure >140/90
• BMI in diabetes population tested
• BMI in diabetes population => 25
• BMI in diabetes population => 30
• Smoking in diabetes population

All of these factors have a well defined normal value and a cut off point where intervention is agreed upon. The consensus is that all of these risk factors should be measured at least annually. Most of the data originate from clinical databases, as national or regional samples that were more or less representative for the national population. The complete set of indicators can be found in the final report of the EUCID project.

As examples for EUPHIX, data are reproduced on the proportion of the diabetics tested that have a high blood pressure, are overweight, and smoke.

Hypertension prevalence in diabetics ranges from 17% to 46%

The upper limit of an acceptable blood pressure is different in different guidelines. The value has decreased over the years. The acceptable limit of 140/90 mm Hg is the highest in use at the moment. This threshold was selected to avoid discussion between the supporters of the different guidelines.

12 Countries were able to provide data for this indicator in the EUCID project, as shown by the Crude percentages of the diabetic population with blood pressure measured in the last 12 months and above 140/90 mm Hg. The percentage of the diabetic population with a blood pressure above the selected threshold varies from 17% in France to 46% in Sweden. In all countries, there is a clear age effect, with the percentages varying from 10-20% around age 30, to 30-80% at age 85.

For more information on the occurrence, prevention, treatment and consequences of hypertension see the EUphact Blood Pressure.

A related set of data from EUCID shows that between 45% (Ireland) and 99% (the Netherlands) of diabetics are regularly tested for their total cholesterol levels.

In diabetics, the prevalence of overweight, including obesity, ranges from 60 to 82%

Overweight and especially obesity are risk factors for macrovascular disease as well as modifiers for the effectiveness of blood glucose treatment. All the guidelines have included BMI (Body Mass Index) as a measurement that should be performed at least once a year. BMI is defined as body weight divided by the square of body length (kg/m2). Normally, 4 categories are used for BMI values:

• below 20.0: underweight
• 20.0 to 24.9: optimal weight
• 25.0 to 29.9: overweight
• 30.0 and above: obesity

The ideal BMI, also for the diabetes population, is between 20 and 25.

12 Countries were able to provide this indicator in the EUCID project. As can be seen in the Crude percentages of overweight and obesity in the diabetic population > 25 years, the majority of the diabetic population has a BMI above 25. The percentages of the diabetic population with a BMI above 25 vary from 59% to 83%. In most countries the percentage of the diabetic population with a BMI above 30 is a little higher than the percentage with a BMI between 25 and 30. These values found for the diabetic population are substantially higher than in the general population, in almost all the countries.

For more information on the occurrence, prevention, treatment and consequences of overweight see the EUphact Overweight.

Smoking in diabetics ranges from 10-37%

Smoking is a risk factor for macrovascular disease. All guidelines advise people with diabetes to stop smoking.

11 Countries were able to provide this indicator in the EUCID project. As shown in the Crude percentages of current smokers in the diabetic population > 25 years, the percentage of diabetics who smoke varies from 10% (Ireland) to 37% (Denmark). Important variations between countries are observed which may reflect the impact of various public health policies. On average, there are no striking differences when compared to the general population. However, it is important to note that the EUCID project and the EUphact Smoking rely on different definitions for smoking and on different data sources. Data gathering has been difficult for this indicator, which might influence the reliability of the data since the data relied upon self-reporting of the people with diabetes and clinical databases are not always well updated on this item.

Two example from EUCID data: Eye examination and prevalence of dyalisis or renal transplantation

This section focuses on data collected on the complications of diabetes. In total, eleven indicators are included in this EUCID category:

• Fundus tested
• Retinopathy
• Retinopathy and timely laser treatment
• Incidence of blindness
• Creatinine tested
• ESRF (end-stage renal failure)
• Incidence of dialysis and transplantation
• Prevalence of dialysis and transplantation
• Incidence of stroke
• Incidence of myocardial infarction
• Incidence of major amputation

These indicators provide data on the complications of diabetes. While the aforementioned indicators were concentrated on prevention of these complications, these indicators give the end points of diabetes care for several organ systems, including both process and outcome indicators. The complete set of indicators can be found in the final report of the EUCID project.

Two examples from the EUCID project are presented here: the regular eye examination (fundus test) for the prevention of diabetic blindness (process indicator), and the prevalence of dialysis/transplantation in diabetics (outcome indicator). .

Percentage of diabetic population that have had their eye fundus inspected in last 12 months

10 countries were able to provide this indicator in the EUCID project, as shown in The crude percentages of the diabetic population with eye fundus inspection during the last 12 months.

Annual inspection of the eye fundus is advised in most guidelines. Some guidelines advise an inspection of the fundus once every two years if the fundus is normal and there are no other risk factors for the development of retinopathy. Whenever symptoms of decreasing vision or vision defects develop, the damage on the fundus is already considerable and treatment is often too late. Therefore, screening in a phase where there are no symptoms is the only way to diagnose diabetic retinopathy in time.

The percentage of the diabetic population that have had their eye fundus inspected in the last 12 months varies widely from 12 (Ireland) to 84 (The Netherlands).

The prevalence of dialysis or renal transplantation in diabetics ranges from 3 to over 600 per 100,000.

Renal failure is one of the microvascular complications of diabetes. Its treatment by dialysis or renal transplantation is dependent on the availability of facilities and on the policy to dialyse patients with diabetes and end-stage-renal-failure.

11 Countries were able to provide data on this indicator, as seen in the Prevalence of dialysis/transplantation per 100,000 diabetic population. The data show large variation between countries, which may be caused by differences in facilities or policies, but also by differences between the data sources.

Diabetes

Overweight

Blood pressure

Smoking

EUCID EUropean Core Indicators in Diabetes

Cholesterol 5>mmol/ll in the diabetic population aged >25, crude percentage

Crude percentage of diabetic population aged >25 years with blood pressure measured in the last 12 months and above 140/90 mm Hg

Crude prevalence of overweight and obesity in general population aged 25-74 years

Standardised prevalence of overweight and obesity in general population, aged 25-74 years

HbA1C > 7,0% in diabetic population aged >25 years

Standardised annual incidence rates of diabetic /100.000 population 0-14 years

Crude annual incidence rates of diabetes /100.000 population 0-14 years

Prevalence of diagnosed diabetes in general population, all ages, age-standardised per 1000 individuals

Prevalence of diagnosed diabetes in general population, all ages per 1000 individuals

Crude percentage of overweight and obesity in diabetic population >25 years with BMI tested in the last 12 months

Crude percentage of current smokers in diabetic population >25 years

Crude percentage of diabetic population> 25 yr with eye fundus inspection in last 12 months

Standardised Mortality Rate in selected countries, per 100.000 population, 2004-2005

Prevalence of dialysis/transplantation per 100,000 diabetic population.

The crude percentages of the diabetic population with eye fundus inspection during the last 12 months.

Cholesterol > 5 mmol/l in the diabetic population aged >25; crude percentage

The data shown above were collected by the EUCID project for 2005 (2004-2006).

The data presented represent the percentages of the tested diabetic population that show total cholesterol levels > 5 mmol/l.

The data are derived from five different sources:

• National surveys
• Clinical databases on primary care
• Clinical databases with a mix of primary and secondary care
• Clinical databases on secondary care
• Administrative database

The data shown above were collected by the EUCID project for 2005 (2004-2006).

The data presented represent the percentages of the tested diabetic population that show blood pressures above 140/90 mm Hg.

The data are derived from five different sources:

• National surveys
• Clinical databases on primary care
• Clinical databases with a mix of primary and secondary care
• Clinical databases on secondary care
• Administrative database

Also see the EUphact Blood Pressure for more data on hypertension prevalence.

Crude prevalence of overweight and obesity in the general population aged 25-74 yrs

The data shown above were collected by the EUCID project for the year 2005 (+/- 1 year). The crude rates are shown as percentages.

Body Mass Index (BMI) is defined as body weight divided by the square of body length (kg/m2). Data are shown separately for BMI 25.0-29.9 and BMI >= 30.0.

The data originates from Health Interview Surveys (self-reported) or Health Examination Surveys (actually measured). In some investigations self-reporting shows an underestimation, when compared to actual measurement.

For more data on the prevalence of overweight see the EUphact Overweight.

Standardised prevalence of overweight and obesity in the general population aged 25-74 yrs

The data shown above were collected by the EUCID project for the year 2005 (+/- 1 year). Standardization was carried out using the European Standard Population (IARC 1976). This makes data more comparable between countries of differing population age structure.

Body Mass Index (BMI) is defined as body weight divided by the square of body length (kg/m2). Data are shown separately for BMI 25.0-29.9 and BMI >= 30.0.

The data originates from Health Interview Surveys (self-reported) or Health Examination Surveys (actually measured). In some investigations self-reporting shows an underestimation when compared to actual measurement.

For more information on the prevalence of overweight see the EUphact Overweight.

HbA1C > 7.0%, in diabetic population aged >25 years, crude percentages

The data shown above were collected by the EUCID project for 2005 (2004-2006).

The data presented represent the percentages of the tested diabetic population that show HbA1c levels > 7.0%.

The data are derived from four different sources:

• National surveys
• Clinical databases on primary care
• Clinical databases with a mix of primary and secondary care
• Clinical databases on secondary care

Standardised annual incidence rate of diabetes /100.000 population 0-14 years

The data shown above were collected by the EUCID project for the year 2005 (+/- 1 year). Standardization was carried out using the European Standard Population (IARC 1976). This makes data more comparable between countries of differing population age structure.

As indicated in the figure, the data may refer to diabetes type 1 only, type 1 and 2 specified or the two types together. For countries marked ‘#’ there were no or very few children with type 2 included. '*'(Germany) indicates that only data about type 1 are available.

Data originated from 2 different sources:

• The clinical databases covered the national population of the country or complete regions;
• The admistrative databases were either driven by drug use or insurance data.

Crude annual incidence rate of diabetes /100.000 population 0-14 years

The crude data shown above were collected by the EUCID project for the year 2005 (+/- 1 year). As indicated in the figure, the data may refer to diabetes type 1 only, type 1 and 2 specified or the two types together. For countries marked ‘#’ there were no or very few children with type 2 included. '*' (Germany) indicates that only data about type 1 are available.

Data originated from 2 different sources:

• The clinical databases covered the national population of the country or complete regions;
• The admistrative databases were either driven by drug use or insurance data.

The data shown above were collected by the EUCID project for the year 2005 (+/- 1 year). Standardization was carried out using the European Standard Population (IARC 1976). This makes data more comparable between countries of differing population age structure.

Data originated from 3 different sources:

• Health interview surveys: telephone interviews of a representative sample of the national population, excluding people with undiagnosed diabetes.
• Medical records: including all diagnosed persons diabetes patients, who have either been pharmacologically treated or have received dietery advice.
• Administrative databases: including only pharmacologically treated individuals.

Only the second option gives a complete figure for the diagnosed diabetic population. The first option is part of the health interview survey that is regularly carried out in most EU countries.

Prevalence of diagnosed diabetes in the general population, all ages, per 1000 individuals

The non-standardized crude data shown above were collected by the EUCID project for the year 2005 (+/- 1 year).

Data originated from 3 different sources:

• Health interview surveys: telephone interviews of a representative sample of the national population, excluding people with undiagnosed diabetes.
• Medical records: including all diagnosed diabetes patients, who have either been pharmacologically treated or have received dietary advice.
• Administrative databases: including only pharmacologically treated individuals.

Only the second option gives a complete figure for the diagnosed diabetic population. The first option is part of the health interview survey that is regularly carried out in most EU countries.

The data shown above were collected by the EUCID project for 2005 (2004-2006).

The data presented represent the percentages of the tested diabetic population with BMI values of 25.0 - 29.9 (overweight) or >30.0 (obesity). BMI (Body Mass Index) is defined as body weight divided by the square of body length (kg/m2).

The data are derived from five different sources:

• National surveys
• Clinical databases on primary care
• Clinical databases with a mix of primary and secondary care
• Clinical databases on secondary care
• Administrative database

Also see the EUphact Overweight for more data on overweight prevalence.

Crude percentages of current smokers in diabetic population > 25 yr (smoking is defined as any smoking)

The data shown above were collected by the EUCID project for 2005 (2004-2006).

The data presented represent the percentages of smokers, including all types of smokers.

The data are derived from four different sources:

• National surveys
• Clinical databases on primary care
• Clinical databases with a mix of primary and secondary care
• Clinical databases on secondary care

Also see the EUphact Smoking for more data on smoking prevalence.

Crude percentage of the diabetic population> 25 yr with eye fundus inspection in last 12 months

The data shown above were collected by the EUCID project for 2005 (2006). The data represent the crude percentages of the diabetic population > 25 year of age with inspection of the eye fundus during the last 12 months.

The data are derived from five different sources:

• National surveys
• Clinical databases on primary care
• Clinical databases with a mix of primary and secondary care
• Clinical databases on secondary care
• Administrative database

The data shown above were collected by the EUCID project for 2005 (2004-2006).

The data represent the number of all cases with dialysis or renal transplantation per 100,000 diabetic population.

The data are derived from three different sources:

• registers;
• clinical databases;
• administrative database.

Standardised Death Rates for Diabetes, in selected countries, per 100,000 population, 2004-2005

The data shown above were collected by the EUCID project for the years 2004 and 2005. Standardization was carried out using the European Standard Population (IARC 1976). This makes data more comparable between countries of differing population age structure.

The EUCID data include data on mortality for diabetes as primary and/or secondary cause of death (left column). The Eurostat data is restricted to primary causes of death (right column).

List of partners collaborating in EUCID