Identifying avoidable switchboard delays in England's NHS hospitals: phase one of the national SWITCH project.

BACKGROUND: Inter-hospital communication frequently requires mediation via a switchboard. Identifying and eliminating switchboard inefficiencies may improve patient care. METHODS: All 175 acute hospital switchboards in England were contacted six times. Call contents and duration were recorded. No clinician calls or bleeps were connected. RESULTS: The mean delay before contacting a switchboard operative was 55±46 seconds. 115 hospitals (66%) used automated switchboards; 34 of these (30%) had infection control messages. Robot operators introduced an additional 40 second delay versus humans (mean 70.3±28 versus 29.8±23 seconds, p<0.0001). Multivariate analysis identified robot operators (HR 5.1, p<0.0001) and infection control messages (HR 2.9, p=0.003) as predictors of delays over 60 seconds. CONCLUSIONS: There are significant avoidable delays in contacting switchboard operatives across England. Quality improvement is underway.

European larch phenology in the Alps: can we grasp the role of ecological factors by combining field observations and inverse modelling?

Vegetation phenology is strongly influenced by climatic factors. Climate changes may cause phenological variations, especially in the Alps which are considered to be extremely vulnerable to global warming. The main goal of our study is to analyze European larch (Larix decidua Mill.) phenology in alpine environments and the role of the ecological factors involved, using an integrated approach based on accurate field observations and modelling techniques. We present 2 years of field-collected larch phenological data, obtained following a specifically designed observation protocol. We observed that both spring and autumn larch phenology is strongly influenced by altitude. We propose an approach for the optimization of a spring warming model (SW) and the growing season index model (GSI) consisting of a model inversion technique, based on simulated look-up tables (LUTs), that provides robust parameter estimates. The optimized models showed excellent agreement between modelled and observed data: the SW model predicts the beginning of the growing season (B(GS)) with a mean RMSE of 4 days, while GSI gives a prediction of the growing season length (L(GS)) with a RMSE of 5 days. Moreover, we showed that the original GSI parameters led to consistent errors, while the optimized ones significantly increased model accuracy. Finally, we used GSI to investigate interactions of ecological factors during springtime development and autumn senescence. We found that temperature is the most effective factor during spring recovery while photoperiod plays an important role during autumn senescence: photoperiod shows a contrasting effect with altitude decreasing its influence with increasing altitude.