Background: Zoonotic public health issues, such as avian influenza, bovine spongiform encephalopathy or antimicrobial resistance, are complex and research or literature reviews addressing these issues often result in contradictory findings and recommendations. There is increasing momentum to adopt more systematic and explicit approaches when using diverse sources of scientific evidence in zoonotic public health to facilitate evidence-informed decision-making.
Methods: Our research group, comprising of individuals from the Public Health Agency of Canada, and multiple Canadian universities, have completed several independent, but complementary systematic reviews and meta-analyses targeting prioritized zoonotic public health issues on specific intervention, etiology, and diagnostic test questions. For the purposes of this presentation we selected three reviews to illustrate our experience with this methodology and to discuss the main challenges and opportunities for utilizing this method in zoonotic public health. These are: a systematic review investigating the zoonotic potential of Mycobacterium avium ssp. paratuberculosis (MAP) (Waddell et al, submitted); a systematic review investigating the effectiveness of on-farm interventions to reduce the shedding of E. coli O157 in the feces of weaned domestic ruminants (Sargeant et al, submitted); and a systematic review investigating the performance of multiple diagnostic tests for Salmonella spp. in swine (Wilkins et al., in progress).
Results and Discussion: Systematic review methodology has been rarely utilized in zoonotic public health. Our initial experience indicate this method is useful for identification of effective interventions in a transparent manner, qualitative summarization and interpretation of research evidence from multiple study designs, evaluation of major strengths and weaknesses in existing research, knowledge gaps for targeting further research efforts, and generation of evidence-based inputs for risk assessment or risk management planning. Some challenges were common to implementing systematic reviews in zoonotic public health: most of existing systematic review formats are primarily developed for evaluating intervention research. This area covers a wide range of research, which results in complex search strategies and a lack of appropriate methodological filters. The resulting search and relevance screening stages consume a lot of time and manpower. There is a lack of acceptable quality studies either due to poor study design or poor reporting, a general lack of randomized controlled trials and a need to rely on observational studies as the main source of evidence. Due to heterogeneity among studies, meta-synthesis part of meta-analyses has been precluded. Challenges specific to certain questions were also observed. For example, much of the scientific evidence pertaining to 'intervention effectiveness questions' comes from challenge trials (artificial disease challenge). This study design is common in the agri-food sector, but not applicable to human health for ethical reasons. This design often has good internal validity, but due to their experimental setting the results are not representative of field conditions. Systematic reviews of diagnostic test performance are virtually non-existent in zoonotic public health. Our major challenge with this review has been a lack of standardization in the conduct and reporting of the diagnostic tests; laboratories frequently develop their own unique protocols, which questions the external validity of test comparisons. More validated and standardized formats for etiology (risk factor) research are urgently needed because risk factor questions are often relevant to policy and decision makers in zoonotic public health.
Conclusions: Systematic review and meta-analysis are powerful methods for identifying, evaluating and synthesizing research evidence, and have been under utilized in zoonotic public health. The need to use these tools in non-traditional public health areas is being increasingly recognized both by researchers and policy makers. The Cochrane Collaboration should play an important role in developing specific guidelines for applying their existing systematic review format to zoonotic public health.