| dc.description.abstract | The purpose of this study was to assess the intrinsic and contextual data quality of the Dutch DBC Information System (DIS) database, which contains all data relevant to the healthcare reimbursements. The intrinsic data quality was examined from a process point of view by determining the organizational and technical issues negatively influencing the data quality. Focus here was on the Dutch hospitals. The contextual data quality was examined as the "fit for use" of these data to evaluate efficiency and patient safety performances of the hospitals. This is currently unexplored, although traditional datasets used for scientific research of these contexts are no longer properly available and alternatives have to be found.
A qualitative research method was used for this research: eighteen field experts were interviewed in nineteen semi-structured expert interviews. The experts represented several relevant stakeholder groups such as the internal hospital administration, the DIS itself, internal and external hospital consultants and scientists specialized in efficiency or patient safety research. To structure the interviews and analyze the results, a data quality framework was constructed based on existing literature.
The interviews were recorded and transcribed. These transcripts served as the resulting (qualitative) dataset and were analyzed with a grounded-theory approach.
The results of these analyses indicated that several organizational and technical issues exist in the hospitals and the DBC system, negatively influencing the (intrinsic) DIS data quality. Most notably the issues pertain to the registration of the performed activities: the management of the hospital activity codes and aggregation to the national activity (ZA) codes used for DIS. Recommendations are provided to improve these issues and ultimately the DIS data quality. Despite the existing issues, the data are still suited for analysis, if used properly. For the evaluation of efficiency, the data have good potential to analyze the technical efficiency of the production (i.e., provided care). For evaluation of patient safety, data mining approaches can be applied on the data to identify unexpected deviations in the treatment path. This research provides the required business and data understanding for such analyses; the actual performing of these analyses are left for future research. | |
