Persistent postoperative pain and the problem of strictly observational research

Risk factors and prediction modelling for chronic post‐surgical pain after breast cancer surgery

“…Some factors appear to be correlated yet might be a proxy for another known variable. For example, depression and anxiety were observed to be collinear by Dereu et al., who also observed pre‐operative expected pain intensity (>6 in 0–10) to be an independent predictor [10]. Given our knowledge that chronic post‐surgical pain develops as a transition during the peri‐operative period, considering only pre‐operative or a particular phase of surgery does not seem justifiable for a complete model.…”

“…It should be considered appropriate to assess the outcome of interest at its earliest and definitive presence rather than at a much later point, such as at 6 months or more after the initial insult [8,9]. It behoves us to intervene to modify and treat persisting pain, which will then influence future predictions of the same disease state [10]. So, any long-term prediction needs to account for the treatments being introduced and the variability in patient responses to these treatments.…”