Time to onset in statistical signal detection revisited: A follow-up study in long-term onset adverse drug reactions.

PURPOSE: In a previous study, we developed a signal detection method using the time to onset (TTO) of adverse drug reactions (ADRs). The aim of the current study was to investigate this method in a subset of ADRs with a longer TTO and to compare its performance with disproportionality analysis. METHODS: Using The Netherlands's spontaneous reporting database, TTO distributions for drug-ADR associations with a median TTO of 7 days or more were compared with other drugs with the same ADR using the two-sample Anderson-Darling (AD) test. Presence in the Summary of Product Characteristics (SPC) was used as the gold standard for identification of a true ADR. Twelve combinations with different values for the number of reports and median TTO were tested. Performance in terms of sensitivity and positive predictive value (PPV) was compared with disproportionality analysis. A sensitivity analysis was performed to compare the results with those from the previous study. RESULTS: A total of 38 017 case reports, containing 32 478 unique drug-ADR associations. Sensitivity was lower for the TTO method (range 0.08-0.34) compared with disproportionality analysis (range 0.60-0.87), whereas PPV was similar for both methods (range 0.93-1.0). The results from the sensitivity analysis were similar to the original analysis. CONCLUSIONS: Because of its low sensitivity, the developed TTO method cannot replace disproportionality analysis as a signal detection tool. It may be useful in combination with other methods.

A prediction model-based algorithm for computer-assisted database screening of adverse drug reactions in the Netherlands.

PURPOSE: The statistical screening of pharmacovigilance databases containing spontaneously reported adverse drug reactions (ADRs) is mainly based on disproportionality analysis. The aim of this study was to improve the efficiency of full database screening using a prediction model-based approach. METHODS: A logistic regression-based prediction model containing 5 candidate predictors was developed and internally validated using the Summary of Product Characteristics as the gold standard for the outcome. All drug-ADR associations, with the exception of those related to vaccines, with a minimum of 3 reports formed the training data for the model. Performance was based on the area under the receiver operating characteristic curve (AUC). Results were compared with the current method of database screening based on the number of previously analyzed associations. RESULTS: A total of 25 026 unique drug-ADR associations formed the training data for the model. The final model contained all 5 candidate predictors (number of reports, disproportionality, reports from healthcare professionals, reports from marketing authorization holders, Naranjo score). The AUC for the full model was 0.740 (95% CI; 0.734-0.747). The internal validity was good based on the calibration curve and bootstrapping analysis (AUC after bootstrapping = 0.739). Compared with the old method, the AUC increased from 0.649 to 0.740, and the proportion of potential signals increased by approximately 50% (from 12.3% to 19.4%). CONCLUSIONS: A prediction model-based approach can be a useful tool to create priority-based listings for signal detection in databases consisting of spontaneous ADRs.

The value of time-to-onset in statistical signal detection of adverse drug reactions: a comparison with disproportionality analysis in spontaneous reports from the Netherlands.

PURPOSE: In pharmacovigilance, the commonly used disproportionality analysis (DPA) in statistical signal detection is known to have its limitations. The aim of this study was to investigate the value of the time to onset (TTO) of ADRs in addition to DPA. METHODS: We performed a pilot study using individual case safety reports (ICSRs) for three drugs (Cervarix®, nitrofurantoin and simvastatin) from the Lareb spontaneous reporting database. TTO distributions for drug - ADR associations were compared to other ADRs for the same drug and to other drugs for the same ADR using two-sample Anderson-Darling testing. Statistically significant associations were considered true positive (TP) signals if the association was present in the official product information of the drug. Sensitivity and specificity for the TTO method were compared with the DPA method. As a measure of disproportionality, the reporting odds ratio (ROR) was used. RESULTS: In general, sensitivity was lower, and specificity was higher for the TTO method compared to DPA. The TTO method showed similar sensitivity for all three drugs, whereas specificity was lower for Cervarix®. Eight additional TP signals were found using the TTO method compared to DPA. CONCLUSIONS: Our study shows that statistical signal detection based on the TTO alone resulted in a limited number of additional signals compared to DPA. We therefore conclude that the TTO method is of limited value for full database statistical screening in our setting. Copyright © 2016 John Wiley & Sons, Ltd.

Spontaneously reported adverse events following COVID-19 basic and booster immunizations in the Netherlands.

INTRODUCTION: The rapid roll-out of novel COVID-19 vaccines made near real-time post-marketing safety surveillance essential to identify rare and long-term adverse events following immunization (AEFIs). In light of the ongoing booster vaccination campaigns, it is key to monitor changes in observed safety patterns post-vaccination. The effect of sequential COVID-19 vaccinations, as well as heterologous vaccination sequences, on the observed post-vaccination safety pattern, remains largely unknown. METHODS: The primary objective of this study was to describe the profile of spontaneously reported AEFIs following COVID-19 vaccination in the Netherlands, including the primary and booster series. Reports from consumers and healthcare professionals were collected via a COVID-19 vaccine-tailored online reporting form by the National Pharmacovigilance Centre Lareb (Lareb) between 6 January 2021 and 31 August 2022. The data were used to describe the most frequently reported AEFIs per vaccination moment, the consumer experienced burden per AEFI, and differences in AEFIs reported for homologous and heterologous vaccination sequences. RESULTS: Lareb received 227,884 spontaneous reports over a period of twenty months. Overall, a high degree of similarity in local and systemic AEFIs per vaccination moment was observed, with no apparent change in the number of reports of serious adverse events after multiple COVID-19 vaccinations. No differences in the pattern of reported AEFIs per vaccination sequence was observed. CONCLUSION: Spontaneous reported AEFIs demonstrated a similar reporting pattern for homologous and heterologous primary and booster series of COVID-19 vaccination in the Netherlands.

Disease-specific ADRs of TNF-α inhibitors as reported by patients with inflammatory rheumatic diseases: a registry-based prospective multicenter cohort study.

BACKGROUND: The extent to which adverse drug reactions (ADRs) of biologics differ per immune-mediated inflammatory disease (IMID), and the relevance of tailoring ADR information per IMID is not fully investigated. We aimed to compare patient-reported ADRs attributed to adalimumab and etanercept between different inflammatory rheumatic diseases (IRDs). RESEARCH DESIGN AND METHODS: ADR reports from IRD patients were extracted from the Dutch Biologic Monitor. ADR frequencies were compared using Fischer-Freeman-Halton exact test and the influence of covariates was assessed using binomial logistic regression.A total, of 729 participants were included, of which 354 participants reported 887 unique ADRs. ADR frequencies were not significantly different between the IRDs. Rheumatoid arthritis and ankylosing spondylitis including axial spondyloarthritis patients had an increased risk of ADRs related to 'Respiratory, thoracic and mediastinal disorders' and as compared to psoriatic arthritis patients. Etanercept use, combination therapy with methotrexate and/or corticosteroids, and age also influenced the risk of reporting specific ADRs. CONCLUSIONS: There were no differences in frequencies and nature of patient-reported ADRs attributed to adalimumab and etanercept between different IRDs. However, more research is needed to align patients' and health-care professionals' perspectives to improve knowledge on disease-specific ADRs.

Quantification of Adverse Drug Reactions Related to Drug Switches in The Netherlands.

We performed a retrospective cohort study in the Dutch patient population to identify active substances with a relatively high number of adverse drug reactions (ADRs) potentially related to drug switching. For this, we analyzed drug switches and reported ADRs related to switching between June 1, 2009, and December 31, 2016, for a selection of 20 active substances. We also compared pharmacovigilance analyses based on the absolute, switch-corrected, and user-corrected numbers of ADRs. In total, 1,348 reported ADRs and over 23.8 million drug switches were obtained from the National Health Care Institute in The Netherlands and from Lareb, which is The Netherlands Pharmacovigilance Centre. There was no correlation between the number of ADRs and the number of switches, but, on average, we found 5.7 reported ADRs per 100,000 switches. The number was relatively high for rivastigmine, levothyroxine, methylphenidate, and salbutamol, with 74.9, 50.9, 47.6, and 26.1 ADRs per 100,000 switches, respectively. When comparing analyses using the absolute number and the switch-corrected number of ADRs, we demonstrate that different active substances would be identified as having a relatively high number of ADRs, and different time periods of increased numbers of ADRs would be observed. We also demonstrate similar results when using the user-corrected number of ADRs instead of the switch-corrected number of ADRs, allowing for a more feasible approach in pharmacovigilance practice. This study demonstrates that pharmacovigilance analyses of switch-related ADRs leads to different results when the number of reported ADRs is corrected for the actual number of drug switches.

When More Is Less: An Exploratory Study of the Precautionary Reporting Bias and Its Impact on Safety Signal Detection.

Concerns have been expressed that large numbers of nonvalue-added reports have been accumulating in adverse drug reaction (ADR) databases, for example, via patient support programs. We performed an assessment of the impact of such reports, which we refer to as "precautionary reports," on safety signal detection in the Netherlands. The case narratives of ADR reports of three case products were screened with text-mining algorithms to identify those reports that lack a causal relationship with the suspected medicinal product. We demonstrate that precautionary reports impede the optimal use of the pharmacovigilance system by, on the one hand, masking safety signals and, on the other hand, creating spurious signals. The precautionary reporting bias and its suppressing effect on statistical signal detection results in an altered adverse event safety profile. The findings from this study highlight the need for a better alignment between regulatory authorities and marketing authorization holders regarding pharmacovigilance guidelines.

Traceability of Biologics in The Netherlands: An Analysis of Information-Recording Systems in Clinical Practice and Spontaneous ADR Reports.

INTRODUCTION AND OBJECTIVE: Pharmacovigilance requirements for biologics mandate that EU Member States shall ensure that any biologic that is the subject of a suspected adverse drug reaction (ADR) is identifiable by brand name and batch number. Recent studies showed that brand name identification is well established, whereas batch numbers are (still) poorly reported. We evaluated information-recording systems and practices in the Dutch hospital setting to identify determinants for brand name and batch number recording as well as success factors and bottlenecks for traceability. METHODS: We surveyed Dutch hospital pharmacists with an online questionnaire on systems and practices in hospitals for recording brand names and batch numbers. Additionally, we performed an analysis of the traceability of recombinant biologics in spontaneous ADR reports (received between 2009 and 2014) from the Netherlands Pharmacovigilance Centre Lareb. RESULTS: The survey showed that brand names are not routinely recorded in the clinical practice of Dutch hospitals, whereas batch numbers are poorly recorded. Seventy-six percent of the 1523 ADR reports for recombinant biologics had a traceable brand name whereas 5% of these reports contained a batch number. The results suggest a possible relationship between the availability of brand and batch number information in clinical practice and the inclusion of this information in ADR reports for biologics. CONCLUSION: The limited traceability of brand names and batch numbers in ADR reports may be primarily caused by the shortcomings in the recording of information in clinical practice. We recommend efforts to improve information-recording systems as a first step to improve the traceability of biologics in ADR reporting.

Parametric time-to-onset models were developed to improve causality assessment of adverse drug reactions from antidiabetic drugs.

OBJECTIVES: The aim of this study was to investigate whether the time to onset (TTO) of common adverse drug reactions (ADRs) of antidiabetic drugs could be modeled using parametric distributions and whether these TTO distributions were dependent on patient characteristics. Furthermore, information relevant for daily clinical practice was to be obtained. STUDY DESIGN AND SETTING: We performed an exploratory TTO modeling study, using a cohort of diabetes mellitus patients. Four parametric distributions (exponential, lognormal, gamma, and Weibull) were compared in terms of their goodness of fit. Covariates that could influence the TTO were investigated. In addition, TTO mean and median values were summarized for use in clinical practice. RESULTS: Overall, the gamma distribution provided the best goodness of fit, although differences with the Weibull distribution were negligible in some instances. No differences in TTO distributions between different antidiabetic drugs for a given ADR were found. The TTO was influenced by suspected concomitant medication for metformin-associated diarrhea. Mean and median TTO values were similar for different drug-ADR combinations. CONCLUSION: Our study shows that the TTO of common ADRs associated with antidiabetic drugs can be modeled using the gamma or Weibull distribution. Furthermore, clinically relevant information about these ADRs can be obtained.

Hearing impairment associated with oral terbinafine use: a case series and case/non-case analysis in the Netherlands Pharmacovigilance Centre Lareb database and VigiBase™.

BACKGROUND: The Netherlands Pharmacovigilance Centre Lareb received reports of six cases of hearing impairment in association with oral terbinafine use. This study describes these cases and provides support for this association from the Lareb database for spontaneous adverse drug reaction (ADR) reporting and from Vigibase™, the ADR database of the WHO Collaborating Centre for International Drug Monitoring, the Uppsala Monitoring Centre. OBJECTIVES: The objective of the current study was to identify whether the observed association between oral terbinafine use and hearing impairment, based on cases received by Lareb, constitutes a safety signal. METHODS: Cases of hearing impairment in oral terbinafine users are described. In a case/non-case analysis, the strength of the association in Vigibase™ and the Lareb database was determined (date of analysis August 2011) by calculating the reporting odds ratios (RORs), adjusted for possible confounding by age, sex and ototoxic concomitant medication. For the purpose of this study, RORs were calculated for deafness, hypoacusis and the combination of both, defined as hearing impairment. RESULTS: In the Lareb database, six reports concerning individuals aged 31-82 years, who developed hearing impairment after starting oral terbinafine, were present. The use of oral terbinafine was disproportionally associated with hypoacusis in both the Lareb database (adjusted ROR 3.9; 95% CI 1.7, 9.0) and in Vigibase™ (adjusted ROR 1.7; 95% CI 1.0, 2.8). Deafness was not disproportionally present in either of the databases. DISCUSSION: Based on the described cases and the statistical analyses from both databases, a causal relationship between the use of oral terbinafine and hearing impairment is possible. The mechanism by which terbinafine could cause hearing impairment has not been elucidated yet. The pharmacological action of terbinafine is based on the inhibition of squalene epoxidase, an enzyme present in both fungal and human cells. This inhibition might result in a decrease in cholesterol levels in human cells, among which are the outer hair cells of the cochlea. It may be possible that the reduction in cochlear cholesterol levels leads to impaired cochlear function and possibly hearing impairment. CONCLUSION: In this study we describe hearing impairment as a possible ADR of oral terbinafine, based on six case reports and statistical support from Vigibase™ and the Lareb database. To our knowledge this association has not been described before.