Günther Heller1, Rainer Schnell2, Rainer Rossi3, Teresa Thomas4, Rolf F Maier5. 1. Fachbereich Medizin, Philipps-Universität Marburg, Marburg. 2. Lehrstuhl für empirische Sozialforschung, Universität Duisburg-Essen, Duisburg. 3. Kinder- und Jugendmedizin - Perinatalzentrum, Vivantes Klinikum Neukölln, Berlin. 4. Abteilung für Neonatologie, Universitätsklinikum Leipzig, Leipzig. 5. Zentrum für Kinder- und Jugendmedizin, Universitätsklinikum Marburg, Philipps-Universität, Marburg.
Abstract
INTRODUCTION: Numerous studies have investigated volume-outcome relationships in the treatment of very low birth weight infants. However, studies addressing the identification of optimal thresholds when introducing minimum provider volumes for treatment of these infants do not exist. METHODS: Publicly available data (www.perinatalzentren.org) of more than 56,000 infants weighing less than 1250 g at birth (NB<1250) and treated in level-1 perinatal centers (highest level in Germany) between 2010 and 2018 was used for statistical analysis. Potentially avoidable deaths after the introduction of minimum provider volumes were calculated by deducting observed deaths from estimated deaths based on logistic regression models for every existing empirical provider volume. Various smoothing functions were used to ascertain optimal thresholds for minimum provider volumes. RESULTS: Independent of the observation period or smoothing technique, the highest number of potentially avoidable deaths was observed for minimum provider volumes of 50-60 NB<1250 per year. Introducing a minimum provider volume of 50 without a transition period would reduce the number of level-1 perinatal centers to a quarter of the current number in Germany. Approximately 60% of NB<1250 would have to be reallocated. CONCLUSION: Analyses of resulting geographical distances are needed in the preparation of minimum provider volumes for treatment of NB<1250 in Germany. Such analyses should include perinatal centers expected to reach minimum provider volumes after subsequent reallocation in the future. Thieme. All rights reserved.
INTRODUCTION: Numerous studies have investigated volume-outcome relationships in the treatment of very low birth weight infants. However, studies addressing the identification of optimal thresholds when introducing minimum provider volumes for treatment of these infants do not exist. METHODS: Publicly available data (www.perinatalzentren.org) of more than 56,000 infants weighing less than 1250 g at birth (NB<1250) and treated in level-1 perinatal centers (highest level in Germany) between 2010 and 2018 was used for statistical analysis. Potentially avoidable deaths after the introduction of minimum provider volumes were calculated by deducting observed deaths from estimated deaths based on logistic regression models for every existing empirical provider volume. Various smoothing functions were used to ascertain optimal thresholds for minimum provider volumes. RESULTS: Independent of the observation period or smoothing technique, the highest number of potentially avoidable deaths was observed for minimum provider volumes of 50-60 NB<1250 per year. Introducing a minimum provider volume of 50 without a transition period would reduce the number of level-1 perinatal centers to a quarter of the current number in Germany. Approximately 60% of NB<1250 would have to be reallocated. CONCLUSION: Analyses of resulting geographical distances are needed in the preparation of minimum provider volumes for treatment of NB<1250 in Germany. Such analyses should include perinatal centers expected to reach minimum provider volumes after subsequent reallocation in the future. Thieme. All rights reserved.