The association between symptoms, activity tracking and weather conditions in patients with arthritis and other chronic pain

几个世纪以来，人们一直推测天气和关节疼痛之间存在联系。然而，数据质量问题使研究人员无法得出任何明确的结论。英国关节炎研究所资助的一个重要的新项目，将利用智能手机的广泛使用来收集关于患者症状，身体活动和天气状况的定期，高质量的数据，并调查关节炎和其他慢性疼痛患者之间的关系。这个博士项目将使用来自多云与疼痛研究的数据：1.开发和测试一个统计模型，用于根据天气条件预测关节炎患者的疼痛2。调查被动监测的身体活动水平和患者报告的疾病之间的关联在第1年和第2年将包括流行病学，生物统计学，统计计算和预测建模的培训。关于天气和关节炎患者症状的文献综述，重点是分析方法，以及环境暴露对健康短期影响的流行病学模型综述。对通过智能手机GPS定位获得的当地天气状况与通过智能手机应用程序输入的患者症状数据之间的关联进行纵向分析。开发基于天气的疼痛预测模型（“疼痛预测”）。第2年和第3年将包括从链接的活动监测器获取、清理和准备身体活动数据，并确定最适合分析的措施。活动测量和症状数据的纵向分析。基于天气预报的未来关节疼痛预测模型将允许患者计划他们的活动。了解被动监测的体力活动和疾病严重程度之间的关联，将允许开发一种新的结果测量用于临床护理和研究。

For centuries, it has been speculated that there is an association between the weather and joint pain. However, data quality issues have left researchers as yet unable to reach any definite conclusions. Cloudy with a Chance of Pain is a major new project funded by Arthritis Research UK that will exploit the widespread uptake of smartphones to gather regular, high quality data on patient symptoms, physical activity and weather conditions and investigate the relationships between them in patients with arthritis and other chronic pain.This PhD project will use data from the Cloudy with a Chance of Pain study to:1. Develop and test a statistical model for forecasting pain in patients with arthritis based on weather conditions2. Investigate the association between passively-monitored physical activity levels and patient-reported symptomsIn years 1 and 2 will include training in epidemiology, biostatistics, statistical computing and prediction modelling. Literature review on weather and symptoms in patients with arthritis, with a focus on analytical methodology, as well as review of epidemiological models of short-term effects of environmental exposures on health. Longitudinal analysis of association between local weather conditions obtained via smartphone GPS location and patient symptom data entered via a smartphone app. Development of a prediction model for pain based on the weather ("pain forecast").Year 2 and 3 will include obtaining, cleaning and preparing physical activity data from linked activity monitors and determine measures best suited for analysis. Longitudinal analysis of activity measures and symptom data. A prediction model for future joint pain based on weather forecasts would allow patients to plan their activities. An understanding of the association between passively monitored physical activity and disease severity would allow development of a novel outcome measure for use in clinical care and research.