How do patients contribute to signal detection? : A retrospective analysis of spontaneous reporting of adverse drug reactions in the UK's Yellow Card Scheme.

BACKGROUND: In 2005, spontaneous reporting of adverse drug reactions (ADRs) to the UK's Yellow Card Scheme (YCS) was extended to include patient reports. Here, we investigate the potential pharmacovigilance impact of patient reporting.
OBJECTIVES: The aim of the study was to investigate the relative contribution of patient reporting to signal detection through disproportionality analysis.
METHODS: Data were analysed from all reports submitted directly to the YCS between October 2005 and September 2007. Three datasets of drug-ADR pairs were created: one for patient reports, one for healthcare professional (HCP) reports and one for all reports combined. The proportional reporting ratio (PRR) method was used to identify signals of disproportionate reporting (SDRs) in each dataset. The number of SDRs identified from patient and HCP reports were compared, as well as the type of ADR and suspect drug involved. A sensitivity analysis was performed to examine how combining the patient and HCP reports may affect the SDRs identified.
RESULTS: Data were received for 5,180 patient and 20,949 HCP reports, relating to 16,566 and 28,775 drug-ADR pairs, respectively, with 4,340 (10.6 %) pairs found in both datasets. A significantly higher proportion of the SDRs identified from HCP reports involved reactions classified as serious by the Medicines and Healthcare products Regulatory Agency (MHRA), compared with patient reports (n = 931, 48.0 % vs. n = 185, 28.5 %), or involved newly marketed drugs (n = 596, 30.7 % vs. n = 71, 10.9 %). The proportion of SDRs assessed as not listed on the Summary of Product Characteristics (SPC) was similar in each group (~15 %, based on a random sample). After combining the patient and HCP reports, 278 (~11 %) of the SDRs identified when each group was analysed separately no longer met the SDR criteria, including 12 potentially serious ADRs not listed on the product's SPC. On the other hand, the combined dataset identified an additional 508 SDRs that were not identified when patient or HCP reports were analysed separately. Approximately 10 % (n = 47) of these additional SDRs were assessed as serious ADRs and were not listed on the product's SPC.
CONCLUSIONS: Although this study is limited to the UK experience, overall, the results suggest that patient reporting may provide a positive complementary contribution to that of HCPs. Patient reporting may make an important contribution to drug safety by identifying different SDRs not identified from HCP reports alone. The combination of reports from patients and HCPs, however, when used for the purposes of signal detection through disproportionality analysis, may result in the loss of some information. One possible strategy is to conduct such analyses using reports from patients and HCPs combined, as well as separately for each group.