prev next front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |review
Firstly, we briefly clarify some terminology regarding risk factors
We define a risk factor as an attribute or exposure which is associated with an increased* probability of a specified outcome (here schizophrenia) but it is not necessarily causal. (*Some risk factors (also called “protective factors” may decrease the probability of a schizophrenia outcome)
We also use the term ‘risk modifying factor’ for factors which appear to operate within the causal chain (in other words they contribute to the outcome of schizophrenia, but may not be sufficient to cause schizophrenia).
We use the terms “risk indicators” (also termed “risk markers”) and “proxies” for variables that may precede the onset of schizophrenia, but are not considered causal.
For example, enlarged ventricles are a risk indicator for schizophrenia, individuals with enlarged ventricles have a higher risk of schizophrenia but enlarged ventricles do not ‘cause’ schizophrenia.
Similarly, season of birth has been shown to be associated with schizophrenia, with a higher risk for schizophrenia for those born in winter or early spring. However, it is thought that such seasonality is a proxy for another causal agent or agents, such as lower winter sunlight or increased winter infection.
Variables that correlate with an outcome, but do not precede the outcome, are sequelae and are not considered further in this presentation.