As part of the recently established continuous health monitoring system in Germany, the first data collection wave of the German Health Interview and Examination Survey for Adults (DEGS1) was conducted by the Robert Koch Institute between November 2008 and December 2011. The DEGS national health survey system is primarily designed to provide nationally representative health interview and examination data for the population 18–79
years of age and to conduct analyses of time trends in population health and disability. Interview and examination data will be collected in periodically repeated survey waves at 8-year intervals
]. By including a large number of persons who already participated in a previous national health interview and examination survey (GNHIES98), DEGS1 combines a national health survey and a survey follow-up study. In order to keep cross-sectional results of DEGS1 representative at the national level, a number of measures were taken. First, to compensate for attrition and aging of the GNHIES98 cohort, a sample of the population 18–79
years was newly drawn from local population registries using a two-stage stratified cluster sampling procedure. Secondly, results from DEGS1 cross-sectional analyses will be weighted by survey-specific weights considering sampling probabilities and non-response for newly drawn participants as well as re-participation probability for GNHIES98 participants.
In DEGS1, a wide range of objective health measures were collected based on highly standardized measurements and tests including laboratory analyses. Automated medication assessment and coding provided detailed and objective information on current medication use. It will hence be possible to verify self-reported health conditions, to identify undetected cases of disease, and to evaluate current disease or risk factor control in the population. High risk groups and health inequalities will be identified based on analyses stratified by sex, age, educational background, income, professional and employment status, and region of residence. Burden of disease will be assessed by analyzing the association of morbidity and comorbidity with functional impairment, subjective health, health-related quality of life, and disability. In order to explore preventive potential, these associations will be further analyzed with respect to personal and environmental resources, e.g. socio-demographic characteristics, health-related behavior, level of perceived social support, health care services utilization.
Results of trend analyses will be used to identify new health risks, to evaluate the effectiveness of health programs and regulations at the population level (e.g. reduction of smoking rates; improved hypertension control), and to validate absolute risk prediction models. Interpretation of time trends will need to consider survey data as well as additional information, such as trends in cause-specific mortality, disease management program enrollment, treatment patterns, ambulatory care sensitive hospital admissions, and nursing care rates. As an example, the prevalence of persons with known diabetes mellitus is likely to have increased over time due to aging of the population and increases in the prevalence of major diabetes risk factors such as obesity. However, improved awareness, treatment and survival may have contributed to increases in prevalence as well. Further insight will be gained from comparative analyses of time trends in the prevalence of persons with known and undetected diabetes.
With respect to time trend analyses, concepts, indicators, instruments and methods of data collection in DEGS1 were kept the same as in prior national health interview and examination surveys, in particular the GNHIES98, as far as possible. There were certain limitations to this. For example, there was a change in methods for measuring blood pressure and some biochemical measures. Analyses of time trends will hence require cross-calibration studies in some cases. Furthermore, several constructs and instruments were newly added in DEGS1. Health indicators have been continuously reviewed and extended in close cooperation with health policy makers and public health expert groups, in order to permit harmonization with regional population-based studies in Germany and with national health surveys from other countries. In order to improve the comparability of results between countries, great efforts have been undertaken at the EU and OECD level to harmonize indicators and instruments and to rigorously standardize the data collection process
]. Areas of ongoing work on indicators particularly relate to health in older age
DEGS1 response rates among newly recruited study participants are considerably lower than response rates in the GNHIES98 (42% vs. 62%). Declining response rates in population-based health surveys have consistently been reported from other European countries over the past decade
]. Selection bias resulting from selective participation of healthier persons is a concern in any population-based survey
]. Survey results may therefore underestimate the overall prevalence of chronic diseases and disability compared to results from claims data
]. In addition, persons unable to provide written consent and those with significant language barriers were excluded from participation in DEGS1. Thus, DEGS1 is likely to underrepresent adults living in institutions. To some extent, this may also be true for persons with an immigration background, even though sample weights encompass adjustment for nationality (German yes/no). Careful non-response analyses will be carried out based on available information from official health statistics and non-response questionnaires with respect to nursing care enrollment and institutionalization, non-German nationality, self-rated health, history of long-standing chronic illness, smoking habits, educational background, and health care services utilization.
About half of the DEGS1 study population already participated in a previous national health survey (GNHIES98). This provides ten- to 12-year follow-up data on nearly 4000 individuals. It will hence be possible to analyze age- and sex-specific individual level changes in many risk factors and health status indicators available from both surveys as presented in Table
]. The DEGS survey panel will continuously grow and timepoints of measurements will be added as DEGS participants who agreed to be re-contacted will be included in subsequent survey interview and examination waves. In between examination survey waves, DEGS participants will be contacted by telephone and postal questionnaires, in order to assess incident health events and changes in health status and health-related behavior. The panel will be continuously followed for vital status and cause-specific death. Participants will be asked for advance written permission for death certificate review, in order to assess cause-specific mortality. Together, this growing data base will facilitate to study trajectories of risk factors, chronic conditions, and health outcomes. In addition, extended follow-up will increase absolute numbers for estimates of disease incidence and prospective analyses using incident health events and total or cause-specific deaths as outcome measures
In summary, data from DEGS1 and subsequent DEGS survey waves will make essential contributions to the surveillance of communicable as well as non-communicable diseases in Germany. Data collected in periodically repeated national health surveys of adults in Germany will be used for continuous health reporting. Findings will be relevant to health policy planning and evaluation, public health research, and health information of the lay public. Survey participants will be followed for changes in health status and health outcomes. Cross-sectional and longitudinal data will be entered in a large health data platform for health and health care services research.