Potential uses of Bayesian networks as tools for synthesis of systematic reviews of complex interventions.

Bayesian networks (BNs) are tools for representing expert knowledge or evidence. They are especially useful for synthesising evidence or belief concerning a complex intervention, assessing the sensitivity of outcomes to different situations or contextual frameworks and framing decision problems that involve alternative types of intervention. Bayesian networks are useful extensions to logic maps when initiating a review or to facilitate synthesis and bridge the gap between evidence acquisition and decision-making. Formal elicitation techniques allow development of BNs on the basis of expert opinion. Such applications are useful alternatives to 'empty' reviews, which identify knowledge gaps but fail to support decision-making. Where review evidence exists, it can inform the development of a BN. We illustrate the construction of a BN using a motivating example that demonstrates how BNs can ensure coherence, transparently structure the problem addressed by a complex intervention and assess sensitivity to context, all of which are critical components of robust reviews of complex interventions. We suggest that BNs should be utilised to routinely synthesise reviews of complex interventions or empty reviews where decisions must be made despite poor evidence.

The repeatability of vegetation classification and mapping.

The mapping of habitats as defined by plant communities is a common component of the planning and monitoring of conservation management. However, there are major concerns about the subjectivity and risk of observer bias in most commonly used plant community mapping protocols. This study provides the first test of the consistency of habitat maps based on the mapping units defined by the National Vegetation Classification (NVC), the most widely used classification of plant communities used for habitat mapping on conservation sites in the UK. Seven surveyors mapped the same upland site within five weeks in summer 2008 and the spatial correspondence of the resulting maps was assessed. The NVC is a hierarchical classification and pair-wise spatial agreement between maps decreased with lower levels of sub-classification. The average area of agreement between maps was 77.6% at the habitat level, 34.2% at the community level and 18.5% at the sub-community level. Spatial disparity in the location of mapped boundaries between vegetation types only made a small contribution to overall differences; the majority of variation between maps was due to discrepancies in classification, with vegetation types containing similar species composition most often confused. Factors relating to surveyor effort (cost, time taken and length of route) were not able to explain the substantial differences between maps. However, the methods used to assign areas to vegetation type did seem to have an effect, with surveyors who relied primarily on their own experience having the highest levels of mean agreement with other maps. The study raises serious concerns with current practice of using the NVC for site description and monitoring/surveillance. Since this is just a single case study, we recommend that further work is carried out with the aim of determining the degree and source of variation between surveyors and how consistency can be increased.