Assessing interventions applied to target populations is a matter of prime interest. Studies are usually undertaken to see whether an alternative intervention is superior (or at least equivalent) to a comparable standard intervention. This is typically achieved by comparing alternative and standard intervention within a given study, and the developed meta-analytic methodology is building on this assumption. Very little work has been delivered when studies only report results on one of the interventions only, but not on both. This is the situation we consider here, and it is motivated by study reports on two surgeries for treatment of asymptomatic antenatally diagnosed congenital lung malformations in young children. Reports are often only available for one of the two, and restricting analysis on those with results on both surgeries will restrict data to 33% of the potential sources. We show in this paper how data sources can be fused and under which condition this fusion will provide valid results. Application to the case study shows the potential gain of the suggested approach in reaching a more conclusive analysis. We argue that studies should best allow within-study comparison, but if only one intervention information is available (for example, as the required surgery expertise for the comparative intervention is not deliverable at the respective site), harnessing one-group information can provide additional insights.

评估针对目标人群的干预措施是首要任务。通常会进行研究，以查看替代干预是否优于（或至少等效）可比的标准干预。这通常通过比较不同的标准和实现干预内一项给定的研究，而发达的荟萃分析方法就是基于这种假设。当研究仅报告一项干预措施的结果，而未报告两项干预措施的结果时，则完成的工作很少。这是我们在此考虑的情况，这是由有关两项手术的研究报告所激发的，该两项手术用于治疗无症状的产前诊断为先天性肺部畸形的幼儿。报告通常仅适用于这两种报告中的一种，并且限制对两种手术均获得结果的分析只能将数据限制在潜在来源的33％。我们在本文中展示了如何融合数据源，以及在何种条件下这种融合将提供有效的结果。在案例研究中的应用表明，该建议方法在进行更结论性分析时具有潜在的收益。