Creating and dissecting causality issues for non-acute or complex diseases

Is there a crisis in medicine and public health with regard to inadequate understanding of causality issues in the modern era? Koch's postulates are well known. They serviced us well when trying to assess whether an infectious agent (which were usually then bacterial or fungal species) caused a single, acute disease. This simple approach frequently fails when dealing with more complex phenomena, such as multi-factorial diseases, illnesses with step-wise development, or most viruses or novel infectious agents. Similarly, whereas many laboratory investigations focus on studying a single pathway in vitro, biology frequently involves feedback loops (positive or negative) and multiple or alternative pathways. Thus, in vivo phenomena may severely deviate from inferences and expectations based on initial research. This may help to explain some of the recent prominent failures of new drugs under development and controlled clinical trials, and discrepancies among results from cross-sectional, case-control and cohort epidemiologic study designs and/or clinical trials. Furthermore, genetic variations may lead to varying degrees of side effects, some of which may be rare and thus only discerned after widespread usage of a drug. The balancing of benefits and harm can be much more complicated for populations than for individuals, and varying indications and warnings can be problematic issues.