This is a repository copy of Effectiveness of a national quality improvement programme to improve survival after emergency abdominal surgery (EPOCH) : a stepped-wedge cluster-randomised trial. Effectiveness of a national quality improvement programme to improve survival after emergency abdominal surgery (EPOCH) : a stepped-wedge cluster-randomised trial. The Lancet. ISSN 0140-6736 https://doi.org/10.1016/S0140-6736(18)32521-2 eprints@whiterose.ac.uk https://eprints.whiterose.ac.uk/ ReuseThis article is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. This licence only allows you to download this work and share it with others as long as you credit the authors, but you can't change the article in any way or use it commercially. More information and the full terms of the licence here: https://creativecommons.org/licenses/ Implications of all the available evidenceDespite the success of some smaller projects, there was no survival benefit from a national quality improvement programme to implement a care pathway for patients undergoing emergency abdominal surgery. To succeed, large national quality improvement programmes need to allow for differences between hospitals and ensure teams have both the time and resources needed to improve patient care.

Choosing the best-performing server for a particular client from a group of replicated proxies is a dificult task. We offer a novel, two-step technique for server selection that chooses a small subset ofBve servers, and isolates testing to that subset for ten days. We present an empirical evaluation of both our method and previously proposed metrics based on traces to 193 commercial proxies. We show that our technique performs better than any of the other metrics we studied -often one to two seconds better for a one-megabytejle -while requiring considerably less work over time. Metrics such as round-trip time and tests using small files usually select servers that are two to three times worse than the best servel: Network-layer metrics such as minimizing router and autonomous system count poorly predict which server provides the best performance. These metrics often select servers with transfer times four to six times that of the best-performing server:

No abstract

We consider the general question of what are Systems Beyond 3G (836). We propose a set of high-level requirements that are desirable in a B3G system and identify some critical enabling technologies necessary for realization of such systems.

The performance of the base Mobile IP handover algorithm for moving the Mobile Node's network layer point of attachment from one subnet to another has been recognized as a potential performance bottleneck for some time. In this paper, we discuss a collection of algorithms that use a link synchronous approach to Mobile IP handover. In the link synchronous approach, information on the progress of switching the link is used to drive handover at the IP level. We present a comprehensive analysis of handover packet drop, and develop analytical models of how the link synchronous algorithms help to mitigate it. We use data from a handover emulator to test the analytical models, and to compare the performance of the different algorithms under a variety of link conditions. Data from implementations on IS-2000 and 802.11b show how the link synchronous algorithms behave on real radio protocols. The results indicate that the link synchronous algorithms can reduce packet loss substantially, with best results possible if the link layer provides information on the move prior to the link switch.Abbreviations: bw r : wireless link bandwidth, in bits per second; bw w : wired link bandwidth between Foreign Agents, in bits per second; c: fraction of time the wireless link is blocked due to partial serialization during prehandover signaling; f success (t,θ P N ): probability density function for successful completion of prehandover signaling as a function of T PN and having parameter vectorθ PN ; LE PN : link layer prehandover notification event; LE LD : link layer link down event; LE LU : link layer link up event; NE MD : network layer movement detected event; NE RC : network layer routing changed event; P success : probability of signaling success; P failure : probability of signaling failure; S WPo : total size of Postreg signaling on wired link (Handover Request/Handover Reply); S RPr : total size of Prereg prehandover signaling on the wireless link (SolPrFAADvert, PRRtAdvert, and ReqRqst); S RPo : size of Solicitation for Foreign Agent Advertisement sent in Post-MIT; T PN : time between prehandover notification and link down; T L2 : time between link down and link up; T MD : time between link up and movement detected; T RC : time between movement detected and routing changed; T lat : latency of wireless link for sending Prereg signaling; T Prereg PS : time for all Prereg prehandover signaling to traverse the old wireless link prior to link handover; T Prereg RC : time for routing change in Prereg, when routing change is initiated prior to link handover; T oSMIP RC : time for routing change in oSMIP, when LE LU and NE MD are collapsed into one event; T TE : tunnel establishment time in reactive Postreg; T IF : time to send a message from one Foreign Agent to the other; Tin PD : time during and after link handover when Mobile Node's incoming packets are dropped; Tin PD : expected value (mean) of Tin PD ; Tout PD : time during and after link handover when Mobile Node's outgoing packets are dropped.