Population surveys: they are periodically repeated in each country and are based on interview or direct clinical examination of the population. Population surveys include:

• Health Interview Surveys – HIS. Present data on self-reported and self-rated health and may not be sufficient to assess CVD morbidity
• Health Examination Surveys – HES. They are based on clinical examinations of population, therefore are extremely expensive.

Data are not always collected through standardized methods, therefore are scarcely comparable.

• Ad hoc CVD surveys and studies. If they are conducted in representative samples and use standardized methods, such as the questionnaire of the London School of Hygiene and Tropical Medicine (LSHTM) to assess angina pectoris, they can provide important information on risk factors and disease prevalence. Precise guidelines exist on description and monitoring of symptoms as well as the use of clinical data and individual factors (height, weight, cholesterol, etc).

Disease-specific registers: these registers are based on record linkage between different sources (mortality and hospital discharge records, General Practitioners’ registers). They cover large samples of the population and their ability to register events depends on the organization of the health system, the diagnostic methods and applied medical care. Incompleteness of records, inaccuracy in event diagnosis and use of non-standardized diagnostic criteria may compromise the quality of register. These registers, to be reliable and the best source of information on morbidity, must meet a set of quality criteria and be able to cover a representative area of the population.

Registers based on administrative data: in some countries registers based on record linkage of mortality and hospital discharge records are available. These registers have existed for many years in a number of Northern countries, where individuals are identified by a personal identification number(ID), thus allowing record linkage of all information sources. These registers are economical, cover the whole country, all age groups and collect a large number of events; they are limited because they are not planned for scientific research and data are not standardized. Changes in the ICD version or the introduction of new diagnostic criteria may have unpredictable effects. These registers can be used when carefully validated.

Hospital discharge records: they provide the number of hospitalizations but do not consider less serious events or count treatments and rehabilitation subsequent to a single event as separate events. Usually, they do not include neither data provided by emergency and ambulance services, nor events occurring in private hospitals and nursing homes; they do not use standardized diagnostic criteria and are very much dependent on the admission criteria of each hospital. They mainly provide information on acute events but are less useful as to chronic conditions.

Cohort longitudinal studies: they enrol large samples of healthy individuals, measure risk factors and observe the development of the disease over time. These studies are very expensive but essential to measure the causal role of risk factors.

General Practitioners’ (GP) networks: in some countries, GPs’ registers may provide important information on chronic conditions and events which do not require hospitalization. Usually, only volunteer GPs have these registers, which seldom use standardized criteria and definitions, therefore data collected need validation.