Post-marketing withdrawal of 462 medicinal products because of adverse drug reactions: a systematic review of the world literature

Abstract

Background: There have been no studies of the patterns of post-marketing withdrawals of medicinal products to which adverse reactions have been attributed. We identified medicinal products that were withdrawn because of adverse drug reactions, examined the evidence to support such withdrawals, and explored the pattern of withdrawals across countries.

Methods: We searched PubMed, Google Scholar, the WHO's database of drugs, the websites of drug regulatory authorities, and textbooks. We included medicinal products withdrawn between 1950 and 2014 and assessed the levels of evidence used in making withdrawal decisions using the criteria of the Oxford Centre for Evidence Based Medicine.

Results: We identified 462 medicinal products that were withdrawn from the market between 1953 and 2013, the most common reason being hepatotoxicity. The supporting evidence in 72 % of cases consisted of anecdotal reports. Only 43 (9.34 %) drugs were withdrawn worldwide and 179 (39 %) were withdrawn in one country only. Withdrawal was significantly less likely in Africa than in other continents (Europe, the Americas, Asia, and Australasia and Oceania). The median interval between the first reported adverse reaction and the year of first withdrawal was 6 years (IQR, 1-15) and the interval did not consistently shorten over time.

Conclusion: There are discrepancies in the patterns of withdrawal of medicinal products from the market when adverse reactions are suspected, and withdrawals are inconsistent across countries. Greater co-ordination among drug regulatory authorities and increased transparency in reporting suspected adverse drug reactions would help improve current decision-making processes.

Fig. 1

Schematic diagram showing process for inclusion of medicinal products withdrawn after approval because of adverse drug reactions

Fig. 2

Launch year versus interval 1 (time lapse between launch year and first reported adverse drug reaction)

Fig. 3

Launch year versus interval 2 (time lapse between launch year and date of the first withdrawal)

Fig. 4

Launch year versus interval 3 (time lapse between the first reported adverse drug reaction and the date of first withdrawal from launch year)

Fig. 5

Interval between first launch and first ADR report (Interval 1) versus time to withdrawal after first ADR report (Interval 3)

Fig. 6

Schematic diagram of the intervals following the launch of a medicinal product. The shortening in Interval 2 is due to a shortening in Interval 1