Benefits and limits of comparative effectiveness studies in evidence-based conservation

Conservation action aims at halting the erosion of biodiversity. Assessing the outcome of a conservation intervention is thus key to improving its efficiency. This is often done by comparing an intervention to a control. Comparative effectiveness studies, on the other hand, compare multiple conservation interventions among each other. In doing so, one can determine which are the most beneficial interventions despite the lack of a control and a formal experimental design. We use an amphibian conservation study to discuss the benefits and limits of this approach. We used the comparative effectiveness approach to evaluate the outcome of a pond creation project. We measured habitat variables at three spatial scales (pond, terrestrial microhabitat, and landscape) and used multistate occupancy and N-mixture models to account for imperfect detection and to relate the explanatory variables to pond colonization, species abundance and the presence of tadpoles (i.e., evidence for successful reproduction). Although characteristics of the created ponds mattered, the availability of suitable terrestrial microhabitat (such as dry stone walls) was even more important in terms of conservation success as measured by colonization and abundance. This case study shows that successful amphibian conservation action depends on landscape complementation, i.e., the paired availability of suitable aquatic and terrestrial microhabitat. We conclude that comparative effectiveness studies can be used to provide critical information for improved conservation action. However, small sample size and a lack of randomization may a priori represent an impediment to strong inference. Nevertheless, comparative effectiveness studies can provide valuable guidance for evidence-based conservation.