Institut für Medizinische Soziologie und Rehabilitationswissenschaft, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Berlin, Deutschland. [Email]
BACKGROUND : Up until now, research data on the implementation of empirical health services research in emergency departments in Germany are scarce. OBJECTIVE : A monitoring instrument applied in a multicenter prospective cohort study in emergency departments (EDs) is described and discussed regarding requirements for the control and supervision of data collection. METHODS : Patients with cardiac diseases, respiratory tract infections, and hip fractures were recruited in eight EDs located in a central district of Berlin. Enrolment figures and nonresponder reasons were analyzed through descriptive statistics. Potential sample bias was examined in terms of response rates as well as the distribution of age and sex in the group of participants and nonresponders. Qualitative content analysis was applied to data from routine supervisory and feedback meetings with study nurses. RESULTS : Within the first 8 months of data collection, 61.1% of the aimed 1104 patients were recruited. Most frequently stated nonresponder reasons were the dense work and care processes in EDs (41.9%) and patients' disease burden (24.7%). Moreover, qualitative results revealed problems with identifying potentially eligible participants and difficulties because of missing research infrastructure in study centers. The response rate of 50.7% and approximately equal distribution of age and sex in participants and nonresponders do not indicate sample biases. CONCLUSIONS : The monitoring instrument has proven to be suited for empirical research in EDs and revealed optimization potential. We recommend using qualitative and quantitative data systematically.