Dan Med J. 2020 Aug 1;67(8):A04190257.
The Danish Cancer Registry (DCR) and the Danish Lung Cancer Registry (DLCR) are nation-wide registries recording Danish patients with lung cancer (LC). The aim of this study was to assess data agreement and possible consequences hereof on estimation of survival between patients in the two registries.
Descriptive statistics were used for comparison of registered patients in 2013-2014 in the DCR and the DLCR. Furthermore, the one-year relative survival (1y-RS) and Cox proportional mortality hazard rates (MRR) were calculated.
In 2013-2014, a total of 9,111 Danish residents were identified with LC in the DCR and 9,316 were found in the DLCR. Merging the two registries showed an agreement of 87%, whereas 6% were included only in the DCR and 8% only in the DLCR. Including patients only registered in one registry, but who seemed to meet the inclusion criteria of both registries, would increase the agreement to 95%. No differences were seen for 1y-RS. However, MRR for patients in the DLCR was significantly lower than for patients in the DCR: 0.94 (95% confidence interval: 0.91-0.98).
Surprisingly, the DCR registered fewer patients in 2013-2014 than the DLCR, even though they employ the same primary data source. The agreement between the DCR and the DLCR was 87%; this may be increased to 95% if patients who seemed to meet the inclusion criteria of the other register were also included. The discrepancies found were mainly due to different definitions of dates of diagnosis, registrations probably missed by the algorithms and possible registration errors. Discrepancies resulted in a significant difference in MRR, but not in 1y-RS.
none.
not relevant.
丹麦癌症登记处（DCR）和丹麦肺癌登记处（DLCR）是记录丹麦肺癌（LC）患者的全国性登记处。本研究的目的是评估这两个登记处之间的数据一致性及其对生存率估计的可能影响。
使用描述性统计比较 2013-2014 年 DCR 和 DLCR 中登记的患者。此外，还计算了一年相对生存率（1y-RS）和 Cox 比例死亡率危险率（MRR）。
2013-2014 年，DCR 共发现 9111 例丹麦居民患有 LC，DLCR 发现 9316 例。合并两个登记处显示一致性为 87%，而 6%仅在 DCR 登记，8%仅在 DLCR 登记。仅将在一个登记处登记的患者纳入，而这些患者似乎符合两个登记处的纳入标准，将使一致性提高到 95%。1y-RS 无差异。然而，DLCR 患者的 MRR 明显低于 DCR 患者：0.94（95%置信区间：0.91-0.98）。
令人惊讶的是，即使使用相同的原始数据来源，DCR 在 2013-2014 年登记的患者也少于 DLCR。DCR 和 DLCR 之间的一致性为 87%；如果将那些似乎符合另一个登记处纳入标准的患者也纳入，则可以提高到 95%。发现的差异主要是由于诊断日期的不同定义、算法可能遗漏的登记以及可能的登记错误。差异导致 MRR 显著差异，但 1y-RS 无差异。
无。
不相关。
