Department of Public Health Solutions, National Institute for Health and Welfare (THL), Helsinki, Finland.
Institute of Public Health and Clinical Nutrition, School of Medicine, Faculty of Health Sciences, University of Eastern Finland, Kuopio, Finland.
Eur J Public Health. 2018 Aug 1;28(4):754-765. doi: 10.1093/eurpub/cky020.
Three data sources are generally used in monitoring health on the population level. Health interview surveys (HISs) are based on participants' self-report. Health examination surveys (HESs) yield more objective data, and also persons who are unaware of their elevated risks can be detected. Medical records (MRs) and other administrative registers also provide objective data, but their availability, coverage and quality vary between countries. We summarized studies comparing self-reported data with (i) measured data from HESs or (ii) MRs. We aimed to describe differences in feasibility and comparability of different data sources for monitoring (i) elevated blood pressure or hypertension (ii) elevated blood glucose or diabetes and (iii) elevated total cholesterol.
We conducted a literature search to identify studies, which validated self-reported measures against objective measures. We found 30 studies published since the year 2000 fulfilling our inclusion criteria (targeted to adults and comparing prevalence among the same persons).
Hypertension and elevated total cholesterol were prone to be under-estimated in HISs. The under-estimate was more pronounced, when the HIS data were compared with HES data, and lower when compared with MRs. For diabetes, the HISs and the objective methods resulted in fairly similar prevalence rates.
The three data sources measure different manifestations of the risk factors and cannot be expected to yield similar prevalence rates. Using HIS data only may lead to under-estimation of elevated risk factor levels or disease prevalence. Whenever possible, information from the three data sources should be evaluated and combined.
通常有三种数据源可用于监测人群健康状况。健康访谈调查（HISs）基于参与者的自我报告。健康检查调查（HESs）可获得更客观的数据，还可以发现那些不知道自己风险增加的人。医疗记录（MRs）和其他行政登记册也提供客观数据，但各国之间可用性、覆盖范围和质量存在差异。我们总结了比较自我报告数据与（i）来自 HESs 的测量数据或（ii）MRs 的研究。我们旨在描述监测（i）血压升高或高血压、（ii）血糖升高或糖尿病和（iii）总胆固醇升高的不同数据源在可行性和可比性方面的差异。
我们进行了文献检索，以确定验证自我报告测量值与客观测量值的研究。我们找到了 30 篇符合我们纳入标准的研究（针对成年人，并比较了同一人群中的患病率），这些研究发表于 2000 年以后。
HISs 容易低估高血压和总胆固醇升高。当将 HIS 数据与 HES 数据进行比较时，低估更为明显，而与 MRs 数据进行比较时则较低。对于糖尿病，HISs 和客观方法得出的患病率相当相似。
三种数据源测量的是风险因素的不同表现，不能期望它们产生相似的患病率。仅使用 HIS 数据可能会导致风险因素水平或疾病患病率的低估。只要有可能，就应评估和结合来自这三种数据源的信息。
