7.1 Introduction to etiological evidence and systematic reviews

In the epidemiological literature, terms such as risk, risk factors, and cause are inconsistently and imprecisely used, and as a result are often misinterpreted leading to incorrect research and policy recommendations (Kraemer, Kazdin et al. 1997). Risk refers to the probability of an outcome within a population of subjects (e.g. risk of lung cancer among people exposed to asbestos) and etiology refers to the cause or the causes (origin) of a certain disease (Kraemer, Kazdin et al. 1997). It is important to distinguish between etiology and risk factors. A risk factor refers to an individual characteristic or exposure that is associated with an increased likelihood of an outcome occurring. For example, are children in sub-Saharan Africa who are exposed to Plasmodium falciparum malaria at an increased risk of developing mental disorders (Akpalu, Ae-Ngibise et al. 2012)? Whereas a protective factor refers to a characteristic or exposure that is associated with the reduced likelihood of an adverse outcome. For example, are people who perform regular higher levels of physical activity less likely to develop lung cancer than those who perform little or no physical activity (Cancer Australia 2014)?

Risk factors are commonly referred to as modifiable, which means they may be controlled or modified in some way, or they may represent a characteristic over which an individual has no control, and therefore categorized as non-modifiable. Exposure to cigarette smoke (either actively or passively), elevated arsenic concentrations, or asbestos in the work or home environment are examples of exposure to modifiable factors – all can ultimately be avoided in most circumstances. Conversely, having a family history of the disease is also known to increase the likelihood of lung cancer development in an individual, and despite any efforts, these non-modifiable risk factors, though less common, are difficult to control or modify (Cancer Australia 2014).

Systematic reviews of etiology and risk factors assess the relationship (association) between certain factors (whether genetic or environmental for example) and the development of a disease or condition or other health outcome. Systematic reviews underpin evidence-based healthcare. The process of conducting a systematic review is a scientific exercise, and as the results will influence healthcare decisions, it is required to have the same rigor expected of all research. The quality of a systematic review depends on the extent to which the methods minimize the risk of error and bias. There is currently no universally accepted methodology for conducting systematic reviews of etiology and risk. Systematic review and meta-analysis of studies related to etiology and risk can provide useful information for healthcare professionals and policymakers on the risk factors (and preventive or protective factors) of disease and where factors, other than direct intervention with therapy and treatment, may influence or impact on health outcomes. Systematic review of etiological studies is important in the public health domain for informing health care planning, resource allocation and strategies for disease prevention.

This chapter outlines and describes JBI's approach and guidance for synthesizing evidence related to etiology and risk and contributes to the emerging field of systematic review methodologies. The systematic review of studies to answer questions of etiology and risk still adheres to the same basic principles of systematic review of other types of data. An a priori protocol must precede and inform the conduct of the systematic review, comprehensive searching must be performed, and critical appraisal of retrieved studies must be carried out followed by data abstraction, analysis and synthesis. These steps will be further discussed in the following sections of this chapter. Additionally, reviewers should refer to two statements/checklists: one for transparent reporting of a systematic review of various research study designs, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Page et al. 2021), and one for Meta-Analyses Of Observational Studies in Epidemiology (MOOSE), which provides a checklist or guidance to report meta-analyses of observational studies in epidemiology, including background, search strategy, methods, results, discussion, and conclusion (Stroup, Berlin et al. 2000).

A note on causation

British epidemiologist Sir Austin Bradford Hill proposed in 1965 a list of nine “viewpoints”, “circumstances” or “aspects” that should be considered when exploring the likelihood of inferring causation from examined associations: strength of the association; consistency of the observed association; specificity of the association; temporal relationship of the association; biological gradient (dose-response); biological plausibility; coherence (cause-effect interpretation of data should not conflict with generally known facts regarding natural history and biology of the disease; experimental evidence; analogy) (Hill 1965). Sir Bradford Hill explicitly stated that none of the nine viewpoints can be used as “indisputable evidence” for or against the causal hypothesis and that these aspects are used to explore more or less likely alternative explanations to the proposed causal explanation for the observed association.

A comprehensive modern discussion about causality (including a critical examination of Hill’s viewpoints) was provided by Rothman et al (2008). It was contended that temporality is a sine qua non for causal explanations of observed associations; however, there is no other criterion other than temporality that is necessary or sufficient criterion for determining whether an observed association is causal (Rothman, Greenland et al. 2008).