The overall aim is to appraise and extend the evidence base on poor quality medicines found in online medicine quality surveys. The thesis starts with two systematic reviews critically appraising the existing evidence, finding a range of evidence gaps subsequently addressed in the following chapters. These include: (a) inconsistent and poor quality reporting, (b) the methods employed are subject to high degrees of risk of bias, (c) discrepancies exist in scientific evidence, and (d) a range of critical therapeutic areas require investigation of the incidence of poor quality medicines. The Delphi consensus study of Chapter 5 constructs the first reporting guidelines to standardise reporting, allowing for evidence-based conclusions. Chapter 6 finds four out of nine investigated withdrawn medicines still available for purchase in a range of countries worldwide (clobenzorex, fenfluramine, rimonabant, and sibutramine), which are subsequently acquired and chemically analysed in Chapter 7. The results from Chapter 8 find a mean of 7.7% poor quality medicines, which corroborates with the findings of the systematic review of Chapter 4, finding a mean of 10% (IQR 0-20%) poor quality medicines in online medicine quality surveys. Purchased medicines displayed a range of quality defects, including orders lacking patient information leaflets, failing pharmacopoeia bioavailability testing, the absence of various types of packaging, the non-delivery of paid orders, and failure of pharmacopoeia friability testing.