Technische Universität München, Uptown München Campus D, Georg-Brauchle-Ring 60/62, 80992, München, Germany.
Institute for Medical Information Processing, Biometry, and Epidemiology (IBE), Ludwig-Maximilians-University Munich, Munich, Germany.
BMC Cardiovasc Disord. 2022 Feb 26;22(1):72. doi: 10.1186/s12872-022-02513-z.
Coronary angiographies (CAs) are among the most common diagnostic procedures carried out in German hospitals, and substantial regional differences in their frequency of use have been documented. Given the heterogeneity with regard to the expected benefits and the varying scope for discretion depending on the indication for the procedure, we hypothesized that the observed variation and the association of need and supply factors differs by indication for CA.
We investigated the correlation between supply factors and the regional rates of CAs in Germany while controlling for need using spatial-autoregressive error models (SARE) and spatial cross-regressive models with autoregressive errors (SCRARE). The overall rates of CAs and the rates in specific patient subgroups, namely, patients with and without myocardial infarction (MI), were calculated based on a comprehensive set of nationwide routine data from three statutory health insurances at the district level.
Although little variation was found in cases with MI, considerable variation was seen in the overall cases and cases without MI. The SARE models revealed a positive association between the number of hospitals with a cardiac catheterization laboratory per 10,000 population and the rates of overall cases and cases without MI, whereas no such relationship existed in cases with MI. Additionally, an association between regional deprivation and the rates of CAs was found in cases with MI, but no such association was seen in cases without MI.
The results supported the hypothesis that the relative association of need and supply factors differed by the indication for CA. Although the regional differences in the frequency of use of CAs can only be explained in part by the factors examined in our study, it offers insight into patient access to and the provision of CA services and can provide a platform for further local research.
冠状动脉造影（CA）是德国医院最常见的诊断程序之一，其使用频率存在显著的地域差异。鉴于预期收益存在异质性，并且根据程序的适应证存在不同的酌处权范围，我们假设观察到的变异以及需求和供应因素的相关性因 CA 的适应证而异。
我们使用空间自回归误差模型（SARE）和带有自回归误差的空间交叉回归模型（SCRARE），在控制需求的情况下，研究了供应因素与德国 CA 区域率之间的相关性。我们根据三个法定健康保险在地区层面的综合全国常规数据，计算了 CA 的总体率和特定患者亚组（即有或无心肌梗死 [MI] 的患者）的比率。
尽管 MI 患者的病例中变异较小，但总体病例和无 MI 的病例中存在相当大的变异。SARE 模型显示，每 10,000 人口中具有心脏导管插入术实验室的医院数量与总体病例和无 MI 病例的比率之间呈正相关，而 MI 患者中不存在这种关系。此外，MI 患者的 CA 率与区域贫困之间存在关联，但无 MI 患者的 CA 率则没有这种关联。
结果支持了这样的假设，即需求和供应因素的相对关联因 CA 的适应证而异。尽管我们研究中检查的因素只能部分解释 CA 使用频率的地域差异，但它提供了对患者获得 CA 服务和 CA 服务提供的深入了解，并为进一步的本地研究提供了平台。
