Causal reasoning in medicine is a frequent part of clinical decision-making, reading research, and making recommendations, but it is not easy and requires special attention. The authors emphasize that the field of causal estimation from data (causal inference) has modern roots only in the beginning of the 20th century, and in recent decades its importance has increased in health research. A common problem is the complexity of the "causal web," where many variables can bias estimates of causal effects and lead to erroneous conclusions. Progress in evaluating the effect of interventions cannot be achieved by advanced statistical analysis alone; explicit assumptions about how the world works and an assessment of whether these assumptions are met are required. The authors caution that without considering these assumptions, effect estimates may be misleading. Therefore, drawing reliable conclusions about causal effects requires a combination of theoretically justified assumptions and careful empirical analysis.