Late presentation for kidney replacement therapy (KRT) is an important cause of avoidable morbidity and mortality. Here, we evaluated the effect of a complex intervention of graphical estimated glomerular filtration rate (eGFR) surveillance across 15% of the United Kingdom population on the rate of late presentation using data routinely collected by the United Kingdom Renal Registry. A stepped wedge cluster randomized trial was established across 19 sites with eGFR graphs generated from all routine blood tests (community and hospital) across the population served by each site. Graphs were reviewed by trained laboratory or clinical staff and high-risk graphs reported to family doctors. Due to delays outside the control of clinicians and researchers few laboratories activated the intervention in their randomly assigned time period, so the trial was converted to a quasi-experimental design. We studied 6,100 kidney failure events at 20 laboratories served by 17 main kidney units. A total of 63,981 graphs were sent out. After adjustment for calendar time there was no significant reduction in the rate of presentation during the intervention period. Therefore, implementation of eGFR graph surveillance did not reduce the rate of late presentation for KRT after adjustment for secular trends. Thus, graphical surveillance is an intervention aimed at reducing late presentation, but more evidence is required before adoption of this strategy can be recommended.

中文翻译：

---

肾功能数据图形监测的阶梯式楔形集群随机试验，以减少肾脏替代治疗的晚期表现


肾脏替代治疗（KRT）的延迟出现是可避免的发病率和死亡率的一个重要原因。在这里，我们使用英国肾脏登记处常规收集的数据，评估了对 15% 的英国人口进行图形估计肾小球滤过率 (eGFR) 监测的复杂干预措施对迟发率的影响。在 19 个地点建立了阶梯式楔形集群随机试验，并使用每个地点所服务人群的所有常规血液检查（社区和医院）生成的 eGFR 图。由经过培训的实验室或临床工作人员审查图表，并将高风险图表报告给家庭医生。由于临床医生和研究人员无法控制的延误，很少有实验室在随机分配的时间段内启动干预措施，因此该试验转为准实验设计。我们在 17 个主要肾病单位服务的 20 个实验室中研究了 6,100 起肾衰竭事件。总共发出了 63,981 张图表。调整日历时间后，干预期间的出现率没有显着降低。因此，在调整长期趋势后，实施 eGFR 图形监测并没有降低 KRT 的迟发率。因此，图形监视是一种旨在减少迟发的干预措施，但在建议采用该策略之前需要更多证据。