Centro de Investigação em Saúde Pública, Escola Nacional de Saúde PúblicaPortuguese School of Public HealthUniversidade Nova de LisboaLisboaPortugal; Centre for Health InformaticsAustralian Institute of Health InnovationMacquarie UniversitySydneyAustralia.
NOVA Medical School/ Faculdade de Ciências Médicas Family Medicine Department Universidade Nova de Lisboa Lisboa Portugal.
JMIR Public Health Surveill. 2016 Mar 17;2(1):e12. doi: 10.2196/publichealth.4319. eCollection 2016 Jan-Jun.
Data routinely collected in electronic health records (EHRs) offer a unique opportunity to monitor chronic health conditions in real-time. Geographic information systems (GIS) may be an important complement in the analysis of those data.
The aim of this study was to explore the feasibility of using primary care EHRs and GIS for population care management and public health surveillance of chronic conditions, in Portugal. Specifically, type 2 diabetes was chosen as a case study, and we aimed to map its prevalence and the presence of comorbidities, as well as to identify possible populations at risk for cardiovascular complications.
Cross-sectional study using individual-level data from 514 primary care centers, collected from three different types of EHRs. Data were obtained on adult patients with type 2 diabetes (identified by the International Classification of Primary Care [ICPC-2] code, T90, in the problems list). GISs were used for mapping the prevalence of diabetes and comorbidities (hypertension, dyslipidemia, and obesity) by parish, in the region of Lisbon and Tagus Valley. Descriptive statistics and multivariate logistic regression were used for data analysis.
We identified 205,068 individuals with the diagnosis of type 2 diabetes, corresponding to a prevalence of 5.6% (205,068/3,659,868) in the study population. The mean age of these patients was 67.5 years, and hypertension was present in 71% (144,938/205,068) of all individuals. There was considerable variation in diagnosed comorbidities across parishes. Diabetes patients with concomitant hypertension or dyslipidemia showed higher odds of having been diagnosed with cardiovascular complications, when adjusting for age and gender (hypertension odds ratio [OR] 2.16, confidence interval [CI] 2.10-2.22; dyslipidemia OR 1.57, CI 1.54-1.60).
Individual-level data from EHRs may play an important role in chronic disease surveillance, namely through the use of GIS. Promoting the quality and comprehensiveness of data, namely through patient involvement in their medical records, is crucial to enhance the feasibility and usefulness of this approach.
电子健康记录（EHR）中常规收集的数据为实时监测慢性健康状况提供了独特的机会。地理信息系统（GIS）可能是分析这些数据的重要补充。
本研究旨在探索使用初级保健 EHR 和 GIS 进行葡萄牙人群慢性病管理和公共卫生监测的可行性。具体来说，选择 2 型糖尿病作为案例研究，旨在绘制其流行程度和合并症的存在情况，并确定可能存在心血管并发症风险的人群。
使用来自三种不同类型 EHR 的 514 个初级保健中心的个体水平数据进行横断面研究。数据来自患有 2 型糖尿病的成年患者（通过问题列表中的国际初级保健分类 [ICPC-2] 代码 T90 识别）。GIS 用于按教区绘制里斯本和塔古斯河谷地区糖尿病和合并症（高血压、血脂异常和肥胖）的流行情况。使用描述性统计和多变量逻辑回归进行数据分析。
我们确定了 205068 名患有 2 型糖尿病的患者，在研究人群中患病率为 5.6%（205068/3659868）。这些患者的平均年龄为 67.5 岁，所有患者中有 71%（144938/205068）患有高血压。在不同的教区中，诊断出的合并症存在相当大的差异。患有高血压或血脂异常的合并症的糖尿病患者在调整年龄和性别后，被诊断患有心血管并发症的几率更高（高血压比值比 [OR] 2.16，置信区间 [CI] 2.10-2.22；血脂异常 OR 1.57，CI 1.54-1.60）。
EHR 的个体水平数据可能在慢性病监测中发挥重要作用，特别是通过使用 GIS。通过患者参与他们的医疗记录来提高数据的质量和全面性，对于增强这种方法的可行性和有用性至关重要。
