Comparing outcomes of hemodialysis and peritoneal dialysis patients: consider the pitfalls

Patients with end-stage renal disease have a limited life expectancy. It is primarily for this reason that there is interest in determining whether one form of dialysis therapy provides any survival advantage over another in different subgroups of individuals. Randomized controlled clinical trials are the gold standard in answering such a question. However, in-center hemodialysis and home peritoneal dialysis require profoundly significant but different effects on patients' lifestyle and it is not surprising that most patients don't want to leave that choice to chance and refuse to participate in such clinical trials. Thus, we have to depend upon observational data to compare outcomes with the two different dialysis therapies. Yet such studies are not without pitfalls, particularly those arising from selection bias, confounding by indication, and residual confounding. These pitfalls and potential methods to overcome them are discussed in this paper. Despite such adjustments, it is impossible to determine if the differences in survival, if any, are a direct result of the dialysis modality or from differences in the characteristics of the patients who are treated with these therapies. Juxtaposing this limitation with the profound impact of dialysis therapies on patients' lives, it is prudent to leave the decision of selection of dialysis modality to the patients themselves and healthcare providers should offer decision support for individuals attempting to make such a selection.