Investigating and implementing scalable interventions to prevent diabetic foot ulcers

截肢是一种毁灭性的、可预防的2型糖尿病并发症。随着糖尿病负担的增加，特别是在低收入和中等收入国家，糖尿病护理在许多情况下仍然有限，包括并发症的管理。糖尿病足部溃疡(DFU)可以通过筛查来预防，特别是在高危人群中，通过积极参与足部监测行为。拟议的研究项目将在秘鲁Peruana Cayetano Heredia大学与几个机构合作进行，并将开发和测试预防糖尿病足部溃疡(DFU)的创新方法。该项目有三个工作包(WP)。在WP1，这项研究将1)开发一种人工智能(AI)算法，该算法将使用脚部自拍图像在溃疡发作之前识别体征，2)共同设计脚部自拍mHealth工具以预防DFU，以及3)试点测试脚部自拍mHealth工具以评估AI算法的可用性和准确性。在WP2中，这项研究将评估脚自拍和脚温在预防DFU方面是否比加强的日常护理更有效，以及脚自拍是否不逊于脚温。WP3侧重于与利益攸关方的接触和实施，贯穿项目的整个过程，探讨如何在秘鲁卫生系统中实施这些干预措施。WP3包括详细的流程评估，以了解复杂的干预措施(脚部自拍和脚温测量)在整个项目中为谁工作，以及如何和为什么工作。此外，经济评估将评估干预措施的成本效益。该项目还致力于建设个人和机构能力。首先，加强社区研究的参与，邀请患者组织的代表进入董事会并参加定期的项目会议。第二，积极参与我们更广泛的研究伙伴网络的能力建设，如患者及其家属、倡导组织和政策制定者，以加强将‘活的’证据纳入研究并增加纳入政策的潜力。我们的建议包含了几项务实的创新，这在LMIC环境下是非常必要的。这款基于家庭的、易于使用、低成本的mHealth足部自拍工具促进了脚部自我监控的习惯，并可以通过量身定制的指导和对关注的特征进行早期分类，为DFU的初级预防做出贡献。随着时间的推移捕捉到的脚的照片记录可以提供客观的评估，医疗保健提供者可以前瞻性地使用这些评估。远程医疗解决了在提供面对面慢性护理方面的限制，在新冠肺炎大流行期间，这种限制加剧了这种限制。利用溃疡前阶段的图像来制定预防DFU的策略是新颖的，而人工智能在预防DFU中的使用以前还没有研究过。最后，使用三臂非劣势设计在LMIC环境中测试两种干预措施是一种使用通用试验平台的稳健的方法学方法。该项目的研究结果旨在克服LMIC在提供慢性护理方面的限制以及糖尿病和DFU专家的短缺。该项目的社会影响有赖于阻止或避免脚部溃疡和截肢的道路，从而改善糖尿病患者的生活质量。

Lower limb amputation is a devastating and preventable complication of type 2 diabetes. As the burden of diabetes rises, particularly in low- and middle-income countries (LMICs), diabetes care remains limited in many settings, including the management of complications. Diabetic foot ulcers (DFU) can be prevented with screening, particularly in high-risk individuals, by actively engaging them in foot monitoring behaviours.The proposed research project will take place at the Universidad Peruana Cayetano Heredia (Peru) in collaboration with several institutions and will develop and test an innovative approach for the prevention of diabetic foot ulcers (DFU). The project has three work packages (WP). In WP1, the study will 1) develop an artificial intelligence (AI) algorithm that will identify signs prior to ulcer onset using foot selfie images, 2) co-design a foot selfie mHealth tool for the prevention of DFU, and 3) pilot-test the foot selfie mHealth tool to assess usability and accuracy of the AI algorithm. In WP2, the study will assess whether foot selfie and foot thermometry are more effective than enhanced usual care in preventing DFU and whether the foot selfie is non-inferior to foot thermometry. WP3 is focused on engagement with stakeholders and implementation and runs across the entire duration of the project to explore how these interventions can be implemented in the Peruvian health system. WP3 includes a detailed process evaluation to understand for whom the complex interventions (foot selfie and foot thermometry) worked for, as well as how and why, throughout this project. In addition, the economic evaluation will assess the cost-effectiveness of the interventions. The project is also committed to building individual and institutional capacity. First, enhancing community research engagement, bringing representatives from patient organisations to sit on the board and attend regular project meetings. Second, actively engaging in building the capacity of our wider network of research partners, such as patients and their families, advocacy organisations, and policymakers to enhance the incorporation of 'lived' evidence into research and increase the potential for uptake into policy. Our proposal encompasses several pragmatic innovations, much needed in LMIC settings. This home-based, easy-to-use, low-cost mHealth foot selfie tool promotes the habit of foot self-monitoring and can contribute to the primary prevention of DFU through tailored guidance and early triage of features of concern. The photographic documentation of the feet captured over time can provide objective assessments that can be used prospectively by healthcare providers. Telemedicine addresses the limitations in the provision of in-person chronic care, which have been exacerbated during the COVID-19 pandemic. The use of images at the pre-ulcer stage to develop a strategy for the prevention of DFU is novel and the use of AI in the prevention of DFU has not been previously studied. Finally, testing two interventions in an LMIC setting with a three-arm non-inferiority design is a robust methodological approach that uses a versatile trial platform.The findings from this project are oriented to overcome the limitations in providing chronic care in LMICs and the shortage of diabetes and DFU specialists. The social impact of the project relies on the interruption or avoidance of the path towards foot ulcers and amputation, thus improving the quality of life of people living with diabetes.