In public health, we are moving towards the third generation of public health information systems. The two articles in this issue [1,2] demonstrate how a cross-national European system has been developed as an integrated knowledge system (EUPHIX), and how a national system (Norhealth) can develop along the same lines. The first generation of public health information systems comprised a data system collecting important basic data on a population and the health of a population, such as births and deaths. Over time, the vital and mortality statistics proved to be a very useful and informative tool. Such statistics have existed for hundreds of years, partly to provide the king of a country with information on the power that he possessed through the size of the population, and partly to obtain knowledge about population health. When the discipline of epidemiology emerged, strict concepts and definitions such as mortality rates and incidence rates were developed. The second generation of public health information systems is characterized by combining time-series data and international comparative data with comments related to the data – for example, comparing ageand sex-standardized mortality rates of lung cancer over time and in different European countries. The public health consequence of such comparisons was an increased focus on highmortality countries and attempts to understand causal pathogenic links between risk factors and disease and mortality – and intervention, e.g. with regard to tobacco smoking. A growing interest in measuring health and disease in live populations and outside institutions developed from the middle of the 20th century. Health interview surveys and health examination surveys of nationally representative populations began to appear in the literature, and lately we have been seeing European attempts to standardize such information systems through standardization of the measurement instrument: the questionnaire. The second generation of public health information systems is also characterized by data collection based on a theoretical concept of health and disease rooted in a multifactorial understanding. Health and disease is determined by many factors, including lifestyle, living conditions, genetics, health system interventions and population structure. The World Health Organization (WHO) programme in Europe, ‘‘Health for All by the Year 2000’’, has been very important and instrumental in the development of relevant health indicators measuring health policy targets. Now – at the beginning of the 21st century – we meet the third generation of public health information systems. I use the term ‘‘integrated knowledge system’’ because it integrates data, descriptive and analytical information, and digested and evidencebased knowledge. Electronic data handling, information technology and the Internet have revolutionized the possibilities of creating integrated health information.