E Hemminki1, M Gissler. 1. National Research and Development Centre for Welfare and Health, Health Services Research Unit, Helsinki, Finland.
Abstract
OBJECTIVE: In obstetrics, both area and hospital variation has previously been documented for operative procedures, but not for other treatments or diagnostic procedures. This study investigated the extent of variation in care during pregnancy, at birth and after birth. DESIGN AND METHODS: The variation was studied by hospitals (n = 52) by hospital level. To account for differences in women's background characteristics, logistic regression analyses were done. The data came from the 1991 Finnish Birth Register which records data on all births (n = 64,171). RESULTS: All interventions and care practices studied varied notably from one hospital to another, with the variation usually being greater within hospitals of the same level than between hospitals of different levels. Each of the seven non-operative interventions showed a larger variation than did each of the four operative interventions. Standardisation for mothers' background characteristics did not eliminate this variation. Some interventions, but not all, were clustered in the same hospitals. CONCLUSIONS: Such large variation indicates a need for further studies on the benefits of obstetric interventions and of care practices, and on better application of research results to obstetric care.
OBJECTIVE: In obstetrics, both area and hospital variation has previously been documented for operative procedures, but not for other treatments or diagnostic procedures. This study investigated the extent of variation in care during pregnancy, at birth and after birth. DESIGN AND METHODS: The variation was studied by hospitals (n = 52) by hospital level. To account for differences in women's background characteristics, logistic regression analyses were done. The data came from the 1991 Finnish Birth Register which records data on all births (n = 64,171). RESULTS: All interventions and care practices studied varied notably from one hospital to another, with the variation usually being greater within hospitals of the same level than between hospitals of different levels. Each of the seven non-operative interventions showed a larger variation than did each of the four operative interventions. Standardisation for mothers' background characteristics did not eliminate this variation. Some interventions, but not all, were clustered in the same hospitals. CONCLUSIONS: Such large variation indicates a need for further studies on the benefits of obstetric interventions and of care practices, and on better application of research results to obstetric care.