Division of Urology, University of Arizona College of Medicine, Tucson, Arizona 85724, USA.
Urology. 2013 Jan;81(1):37-42. doi: 10.1016/j.urology.2011.04.024. Epub 2011 Jun 15.
To evaluate the utility of using Internet search trends data to estimate kidney stone occurrence and understand the priorities of patients with kidney stones. Internet search trends data represent a unique resource for monitoring population self-reported illness and health information-seeking behavior.
The Google Insights for Search analysis tool was used to study searches related to kidney stones, with each search term returning a search volume index (SVI) according to the search frequency relative to the total search volume. SVIs for the term, "kidney stones," were compiled by location and time parameters and compared with the published weather and stone prevalence data. Linear regression analysis was performed to determine the association of the search interest score with known epidemiologic variations in kidney stone disease, including latitude, temperature, season, and state. The frequency of the related search terms was categorized by theme and qualitatively analyzed.
The SVI correlated significantly with established kidney stone epidemiologic predictors. The SVI correlated with the state latitude (R-squared=0.25; P<.001), the state mean annual temperature (R-squared=0.24; P<.001), and state combined sex prevalence (R-squared=0.25; P<.001). Female prevalence correlated more strongly than did male prevalence (R-squared=0.37; P<.001, and R-squared=0.17; P=.003, respectively). The national SVI correlated strongly with the average U.S. temperature by month (R-squared=0.54; P=.007). The search term ranking suggested that Internet users are most interested in the diagnosis, followed by etiology, infections, and treatment.
Geographic and temporal variability in kidney stone disease appear to be accurately reflected in Internet search trends data. Internet search trends data might have broader applications for epidemiologic and urologic research.
评估利用互联网搜索趋势数据来估计肾结石发生的情况，并了解肾结石患者的关注重点。互联网搜索趋势数据是监测人群自我报告疾病和健康信息搜索行为的独特资源。
使用 Google Insights for Search 分析工具研究与肾结石相关的搜索，每个搜索词根据相对于总搜索量的搜索频率返回搜索量指数 (SVI)。通过位置和时间参数编译“肾结石”一词的 SVI，并将其与已公布的天气和结石流行数据进行比较。进行线性回归分析，以确定搜索兴趣评分与肾结石疾病的已知流行病学变化之间的关联，包括纬度、温度、季节和州。通过主题对相关搜索词的频率进行分类，并进行定性分析。
SVI 与已确立的肾结石流行病学预测因素显著相关。SVI 与州纬度（R-squared=0.25；P<.001）、州平均年温度（R-squared=0.24；P<.001）和州综合性别流行率（R-squared=0.25；P<.001）相关。女性流行率的相关性强于男性流行率（R-squared=0.37；P<.001 和 R-squared=0.17；P=.003）。全国 SVI 与美国每月平均温度密切相关（R-squared=0.54；P=.007）。搜索词排名表明，互联网用户最感兴趣的是诊断，其次是病因、感染和治疗。
肾结石疾病的地理和时间变化似乎在互联网搜索趋势数据中得到了准确反映。互联网搜索趋势数据可能在流行病学和泌尿科研究中有更广泛的应用。
