Recurrent miscarriage: do professionals adhere to their guidelines

STUDY QUESTION: Is the actual care for recurrent miscarriage in clinical practice in accordance with 23 guideline-based quality indicators? SUMMARY ANSWER: The accordance of actual care with the guidelines was poor and there is evident room for improvement. WHAT IS KNOWN ALREADY: Evidence-based guidelines are important instruments to improve quality of care, but implementation of guidelines is often problematic. STUDY DESIGN, SIZE, DURATION: A retrospective cohort study was performed within a 12-month period (2006) in nine departments of Obstetrics and Gynaecology in the Netherlands. PARTICIPANTS, SETTING, METHODS: Five hundred and thirty women with recurrent miscarriage were included. Actual care was assessed with 23 guideline-based quality indicators (covering diagnostics, therapy and counselling) by calculating per indicator the percentage of women for whom the indicator was followed. Thereafter we did multilevel analyses, to relate the adherence to the indicator to determinants of women, professionals and hospitals. MAIN RESULTS AND THE ROLE OF CHANCE: Homocysteine and antiphospholipid antibodies were determined in 39 and 47%, respectively. Thrombophilia screening (54%) and karyotyping (50%) were offered to women regardless of their underlying risk for inherited thrombophilia or chromosome abnormalities. Higher maternal age at the time of presentation and a lower number of preceding miscarriages were improperly used to decide on diagnostic tests and were both associated with lower guideline adherence by professionals. Professionals with a subspecialization in recurrent miscarriage performed better standard care, i.e. screening for antiphospholipid antibodies and homocysteine, but also showed overuse of diagnostics in women at low risk of inherited thrombophilia. LIMITATIONS, REASONS FOR CAUTION: Retrospective cohort study. WIDER IMPLICATIONS OF THE FINDINGS: Quality indicators used will enable measurement of quality of care. STUDY FUNDING: The study was funded by The Netherlands Organisation for Health Research and Development (ZonMw) (Grant no. 94517005). None of the authors has any conflict of interest to declare.