Patient claims and complaints data for improving patient safety

Purpose - The purpose of this paper is to describe patient complaints and claims data from Swedish databases and assess their value for scientific research and practical health care improvement. Design/methodology/approach - The article first describes previous research into patient claims and similar schemes. It then presents three types of data on patient claims and complaints in Sweden: data generated by the Patient Insurance Fund, the Medical Responsibility Board and the Patients' Advisory Committees and considers methodological issues in using the data. Findings - The databases' value is problems related to spontaneous reporting, which makes it difficult to know how much the data correspond to general injury rates and health care patterns. Another issue is the balance between the size of study materials and the timeliness, e.g. when diagnosis-specific analysis requires data pooling over several years in order to reach adequate case numbers. Adjustment for confounders not present in the databases, e.g. data on hospital case-mix, may add to difficulties using the data in comparative analyses of safety performance Research limitations/implications - The databases' safety analysis and quality improvement value depends on understanding their function, data collection method and their limitations as a source of data about the true incidence and prevalence of injuries and safety problems. Originality/value - This is the first thorough review of the possibilities and limitations associated with the use of claims and complaints data in health care research and improvement.