Enhancing antidepressant safety surveillance: comparative analysis of adverse drug reaction signals in spontaneous reporting and healthcare claims databases

Background/Objective: Spontaneous reporting systems (SRS) such as the Korea Adverse Event Reporting System (KAERS) are limited in their ability to detect adverse drug reaction (ADR) signals due to their limited data on drug use. Conversely, the national health insurance claim (NHIC) data include drug use information for all qualifying residents. This study aimed to compare ADR signal profiles for antidepressants between KAERS and NHIC, evaluating the extent to which detected signals belong to common ADRs and labeling information. Materials and Methods: ADR signal detection in KAERS and NHIC databases, spanning January to December 2017, employed disproportionality analysis. Signal classes were determined based on System Organ Class (SOC) of the Medical Dictionary for Regulatory Activities (MedDRA). Also, Common ADR Coverage (CAC), the proportion of detected signals deemed common ADRs, and labeling information coverage (LIC) represented by mean average precision (mAP) were calculated. Additionally, protopathic bias and relative risk (RR) evaluation were performed to check for signal robustness. Results: Signal detection revealed 51 and 62 signals in KAERS and NHIC databases, respectively. Both systems predominantly captured signals related to nervous system disorders, comprising 33.3% (N = 17) in KAERS and 50.8% (N = 31) in NHIC. Regarding the type of antidepressants, KAERS predominantly reported signals associated with tricyclic antidepressants (TCAs) (N = 21, 41.2%), while NHIC produced most signals linked to selective serotonin reuptake inhibitors (SSRIs) (N = 22, 35.5%). KAERS exhibited higher CAC (68.63% vs. 29.03%) than NHIC. LIC was also higher in KAERS than in NHIC (mAP for EB05: 1.00 vs. 0.983); i.e., NHIC identified 5 signals not documented in drug labeling information, while KAERS found none. Among the unlabeled signals, one (Duloxetine-Myelopathy) was from protopathic bias, and two (duloxetine-myelopathy and tianeptine-osteomalacia) were statistically significant in RR. Conclusion: NHIC exhibited greater capability in detecting ADR signals associated with antidepressant use, encompassing unlabeled ADR signals, compared to KAERS. NHIC also demonstrated greater potential for identifying less common ADRs. Further investigation is needed for signals detected exclusively in NHIC but not covered by labeling information. This study underscores the value of integrating different sources of data, offering substantial regulatory insights and enriching the scope of pharmacovigilance.

Detection of drug safety signal of drug-induced neutropenia and agranulocytosis in all-aged patients using electronic medical records

PURPOSE

We explored the adverse drug reaction signals of drug-induced neutropenia (DIN) and drug-induced agranulocytosis (DIA) in hospitalized patients and evaluated the novelty of these correlations.

METHOD

A two-step method was established to identify the relationship between drugs and DIN or DIA using 5-year electronic medical records (EMRs) obtained from 242 000 patients at Qilu Hospital of Shandong University. First, the drugs suspected to induce DIN or DIA were selected. The associations between suspected drugs and DIN or DIA were evaluated by a retrospective cohort study using unconditional logistic regression analysis and multiple linear regression model.

RESULTS

Twelve suspected drugs (vancomycin, meropenem, voriconazole, acyclovir, ganciclovir, fluconazole, oseltamivir, linezolid, compound borax solution, palonosetron, polyene phosphatidylcholine, and sulfamethoxazole) were associated with DIN, and six suspected drugs (vancomycin, voriconazole, acyclovir, ganciclovir, fluconazole, and oseltamivir) were associated with DIA. The multivariate linear regression model revealed that nine drugs (vancomycin, meropenem, voriconazole, ganciclovir, fluconazole, oseltamivir, compound borax solution, palonosetron, and polyene phosphatidylcholine) and four drugs (vancomycin, voriconazole, ganciclovir, and fluconazole) were found to be associated with DIN and DIA, respectively. While logistic regression analysis revealed that palonosetron and ganciclovir were associated with DIN and DIA, respectively.

CONCLUSION

Palonosetron and ganciclovir were found to be correlated with drug-induced granulocytopenia. The results of this study provide an early warning of drug safety signals for drug-induced granulocytopenia, facilitating a quick and appropriate response for clinicians.

Methods for drug safety signal detection using routinely collected observational electronic health care data: A systematic review

PURPOSE

Signal detection is a crucial step in the discovery of post-marketing adverse drug reactions. There is a growing interest in using routinely collected data to complement established spontaneous report analyses. This work aims to systematically review the methods for drug safety signal detection using routinely collected healthcare data and their performance, both in general and for specific types of drugs and outcomes.

METHODS

We conducted a systematic review following the PRISMA guidelines, and registered a protocol in PROSPERO. MEDLINE, EMBASE, PubMed, Web of Science, Scopus, and the Cochrane Library were searched until July 13, 2021.

RESULTS

The review included 101 articles, among which there were 39 methodological works, 25 performance assessment papers, and 24 observational studies. Methods included adaptations from those used with spontaneous reports, traditional epidemiological designs, methods specific to signal detection with real-world data. More recently, implementations of machine learning have been studied in the literature. Twenty-five studies evaluated method performances, 16 of them using the area under the curve (AUC) for a range of positive and negative controls as their main measure. Despite the likelihood that performance measurement could vary by drug-event pair, only 10 studies reported performance stratified by drugs and outcomes, in a heterogeneous manner. The replicability of the performance assessment results was limited due to lack of transparency in reporting and the lack of a gold standard reference set.

CONCLUSIONS

A variety of methods have been described in the literature for signal detection with routinely collected data. No method showed superior performance in all papers and across all drugs and outcomes, performance assessment and reporting were heterogeneous. However, there is limited evidence that self-controlled designs, high dimensional propensity scores, and machine learning can achieve higher performances than other methods.

Tramadol-induced hiccups: a case-noncase study in the European pharmacovigilance database

Background

Hiccups are usually benign and self-limiting, but can sometimes be persistent. If left untreated, they can provoke severe discomfort, and even death. Hiccups can be idiopathic, organic, psychogenic, and caused by drugs. Although some case reports have suggested a possible association between tramadol and hiccups, to our knowledge, no study has analyzed this possible relationship. The aim of this study was to analyze whether a disproportionate number of cases of hiccups are reported for tramadol in the EudraVigilance database.

Methods

A case-noncase study was conducted to assess the association between hiccups and tramadol, calculating reporting odds ratios (RORs) from 1 January 1995 to 11 September 2020. Cases were selected using the preferred term 'Hiccups'. The noncases used as controls were all other adverse drug reaction reports recorded in EudraVigilance during the same period. Exposure was defined as exposure to tramadol among cases and noncases. To reduce the risk of confounding by indication, the RORs for tramadol compared with other opioids were obtained. Additionally, we performed a confirmatory analysis in the World Health Organization pharmacovigilance database, VigiBase^®.

Results

There were 3089 cases of hiccups in the 7,213,623 reports. Tramadol was involved in 50 cases. The ROR for tramadol exposure was 3.35 [95% confidence interval (CI) 2.53-4.43]. This association persisted when comparing tramadol with other opioids; ROR: 2.13 (95% CI 1.52-2.99). Disproportionality was also observed in VigiBase^®: ROR 1.69 (95% CI 1.47-1.93).

Conclusion

Our study confirms, for the first time, a possible signal for a tramadol-hiccups association. Nevertheless, observational analytical studies are needed to confirm these results.

Plain Language Summary

Evaluation of the relationship between the tramadol and the risk of hiccupsIntroduction: Hiccups are sudden involuntary contractions of the diaphragm. This involuntary contraction causes the vocal cords to close very briefly, which produces the characteristic sound of a hiccup. Hiccups are usually benign and self-limiting, but can sometimes be persistent. If left untreated, they can provoke severe discomfort, depression, disability, and in the most extreme cases, even death. Drugs are a rare cause of hiccups.Methods: This study investigated the possible association between tramadol and hiccups (an unmentioned adverse drug reaction in the Summary of Product Characteristics) in the European pharmacovigilance database (EudraVigilance) and a confirming analysis in the World Health Organization pharmacovigilance database (VigiBase).Results: Our analysis shows that hiccups is relatively more frequently reported in association with tramadol than with other medicinal products, with EudraVigilance and VigiBase confirming this association.Conclusion: Tramadol is an opioid analgesic indicated, alone or in combination with dexketoprofen or paracetamol for pain with various causes, so healthcare professionals and patients should be aware of this possible association.

Using and improving distributed data networks to generate actionable evidence: the case of real-world outcomes in the Food and Drug Administration's Sentinel system

The US Food and Drug Administration (FDA) Sentinel System uses a distributed data network, a common data model, curated real-world data, and distributed analytic tools to generate evidence for FDA decision-making. Sentinel system needs include analytic flexibility, transparency, and reproducibility while protecting patient privacy. Based on over a decade of experience, a critical system limitation is the inability to identify enough medical conditions of interest in observational data to a satisfactory level of accuracy. Improving the system's ability to use computable phenotypes will require an "all of the above" approach that improves use of electronic health data while incorporating the growing array of complementary electronic health record data sources. FDA recently funded a Sentinel System Innovation Center and a Community Building and Outreach Center that will provide a platform for collaboration across disciplines to promote better use of real-world data for decision-making.

A Drug Safety Surveillance Study of a Ciprofloxacin/Dexamethasone Ophthalmic Fixed Combination in Peruvian Population

(1) Background: drugs provide a significant benefit for patients who require medical treatment; however, their use implies an intrinsic potential danger, with the possibility of causing unwanted effects. These effects are known as adverse drug reactions (ADRs). Post-marketing drug safety surveillance detects unknown risks that have not been identified in clinical trials, and it is necessary to monitor marketed medications under real-life practice. Due to the scarce information about fixed combination of ciprofloxacin 0.3%/dexamethasone 0.1% (SDO), we performed a drug safety surveillance study. (2) Methods: A prospective non-controlled drug safety surveillance study was conducted in Peruvian population. A total of 236 patients prescribed SDO were included derived from 12 sites. Patients' standardized information was collected through two phone calls, including demographics, medical history, prescribing patterns of SDO, concomitant medication, and ADRs in detail. The ADRs were classified by causality and severity, followed by outcome measures to identify new risk. (3) Results: 236 patients prescribed with SDO participated in the study and 220 were included. A total of 82 ADRs/220 patients were reported after the use of SDO, presenting a ratio 0.37 ADR/patient. The most frequent ADR with SDO administration was eye irritation (30%). All ADRs were classified as non-serious, and 97.5% (n = 80) were classified as mild while 2.5% as moderate (n = 2). No cases under the severe category were identified. (4) Conclusion: No new risks were found in the population where this study was conducted.

SGLT2 Inhibitors and Bladder Cancer: Analysis of Cases Reported in the European Pharmacovigilance Database

The association between sodium-glucose cotransporter 2 inhibitors (SGLT2is) and cancer risk is unclear. The objective of this study was to analyze whether a disproportionate number of cases of bladder cancer are reported for SGLT2is in EudraVigilance. A case/noncase study was conducted to assess the association between bladder cancer and SGLT2is, calculating reporting odds ratios (RORs) from November 11, 2012 (approval date for the first SGLT2i, dapagliflozin) to May 19, 2020. First, cases involving SGLT2is were compared with those involving all other drugs; and similar analysis was performed for each SGLT2i. Second, to reduce the risk of confounding by indication, the RORs for SGLT2is compared with other antidiabetics were obtained. Besides, 2 measures were taken to evaluate a possible notoriety bias: a sensitivity analysis excluding pioglitazone was performed and the evolution of the ROR over time for SGLT2is was measured. There were 6602 cases of bladder cancer in the 4,213,637 reports during the study period. SGLT2is were involved in 155 cases. The ROR for pooled SGLT2is was 3.97 (95% confidence interval [CI], 3.39-4.66), disproportionality also being observed for each SGLT2i separately. The association was strongest for dapagliflozin (ROR, 7.02; 95%CI, 5.69-8.66). Nonetheless, this association disappeared when comparing SGLT2is with other antidiabetic drugs (ROR, 0.20; 95%CI, 0.17-0.24). But when excluding pioglitazone from the analysis, a safety signal for SGLT2is compared with other antidiabetics emerged (ROR, 6.84; 95%CI 5.41-8.65). Our study found a disproportionately high number of cases of bladder cancer among users of SGLT2is. However, observational analytical studies will be needed to confirm these results.

Comparison of Data Mining Methods for the Signal Detection of Adverse Drug Events with a Hierarchical Structure in Postmarketing Surveillance

There are several different proposed data mining methods for the postmarketing surveillance of drug safety. Adverse events are often classified into a hierarchical structure. Our objective was to compare the performance of several of these different data mining methods for adverse drug events data with a hierarchical structure. We generated datasets based on the World Health Organization's Adverse Reaction Terminology (WHO-ART) hierarchical structure. We evaluated different data mining methods for signal detection, including several frequentist methods such as reporting odds ratio (ROR), proportional reporting ratio (PRR), information component (IC), the likelihood ratio test-based method (LRT), and Bayesian methods such as gamma Poisson shrinker (GPS), Bayesian confidence propagating neural network (BCPNN), the new IC method, and the simplified Bayesian method (sB), as well as the tree-based scan statistic through an extensive simulation study. We also applied the methods to real data on two diabetes drugs, voglibose and acarbose, from the Korea Adverse event reporting system. Only the tree-based scan statistic method maintained the type I error rate at the desired level. Likelihood ratio test-based methods and Bayesian methods tended to be more conservative than other methods in the simulation study and detected fewer signals in the real data example. No method was superior to the others in terms of the statistical power and sensitivity of detecting true signals. It is recommended that those conducting drug‒adverse event surveillance use not just one method, but make a decision based on several methods.

Impact of adverse event reports from marketing authorization holder-sponsored patient support programs on the performance of signal detection in pharmacovigilance

OBJECTIVES

Marketing authorization holder (MAH)-sponsored patient support programs (PSPs) are a major source of adverse event (AE) reports. The impact of reports from PSPs on the ability to detect AE signals is unclear. We compared signal detection performance using data from PSPs vs. non-PSP sources, and between PSPs providing clinical services vs. PSPs not providing clinical services.

METHODS

Data were obtained from an internal safety database for a global pharmaceutical company 2015-2017. We assessed whether signals were detected for the reference drug-AE pairs using data from PSPs vs. non-PSP sources, and among different PSP services. The performance was evaluated by four measures including area under the receiver operating characteristic curve (AUC) and time-to-signal detection.

RESULTS

While the majority of reports were from PSPs, non-PSP sources were better and faster at detecting signals (AUC 0.63 vs. 0.41, p = 0.035; HR 3.52, p = 0.014) compared to PSPs. Within PSPs, PSPs providing clinical services were marginally better at detecting signals (AUC 0.60 vs. 0.41, p = 0.053) but not faster compared to PSPs not providing clinical services.

CONCLUSION

Reports of AEs from PSPs had worse signal detection performance compared to non-PSP sources. Pharmacovigilance experts should be mindful when using databases that contain reports from PSPs for signal detection.

Masking by vaccines in pediatric drug safety signal detection in the EudraVigilance database

PURPOSE

Postmarketing drug safety surveillance relies upon measures of disproportionate reporting in spontaneous reporting systems. It has been hypothesized that products or events reported frequently may "mask" signals.

METHODS

We analyzed the masking effect of vaccines in pediatrics in the EudraVigilance database by conducting disproportionality analysis in the full database (containing vaccine exposures) and in a restricted set (excluding vaccine exposures). We measured performance of the reporting odds ratio (ROR) in both data sets using a pediatric-specific drug reference set and in the absence of a reference set. We assessed masking effects across age groups and conducted a classification tree (CART) analysis.

RESULTS

Removal of vaccines decreased the ROR values both in negative and positive controls. Exceptions were drug-event combinations including outcomes frequent in vaccine reports. When restricted to positive control associations, removal of vaccine-related events resulted in increased ROR values for events commonly reported following vaccination. For events rarely associated with vaccination, ROR values decreased for all age groups, especially infants. Analysis in the absence of a reference set showed decrease in ROR following vaccine removal and CART revealed that change in ROR with vaccine removal depended upon age and proportion of reports including a vaccine.

CONCLUSIONS

Removal of vaccines for signal detection in a pediatric population has an impact on ROR, dependent upon the reporting frequency of the event of interest in combination with vaccines. We recommend stratification by age and removal of vaccine exposures if the investigated adverse drug reactions include those typically reported in association with vaccines for the age strata.

Analyzing Information Seeking and Drug-Safety Alert Response by Health Care Professionals as New Methods for Surveillance

Background

Patterns in general consumer online search logs have been used to monitor health conditions and to predict health-related activities, but the multiple contexts within which consumers perform online searches make significant associations difficult to interpret. Physician information-seeking behavior has typically been analyzed through survey-based approaches and literature reviews. Activity logs from health care professionals using online medical information resources are thus a valuable yet relatively untapped resource for large-scale medical surveillance.

Objective

To analyze health care professionals’ information-seeking behavior and assess the feasibility of measuring drug-safety alert response from the usage logs of an online medical information resource.

Methods

Using two years (2011-2012) of usage logs from UpToDate, we measured the volume of searches related to medical conditions with significant burden in the United States, as well as the seasonal distribution of those searches. We quantified the relationship between searches and resulting page views. Using a large collection of online mainstream media articles and Web log posts we also characterized the uptake of a Food and Drug Administration (FDA) alert via changes in UpToDate search activity compared with general online media activity related to the subject of the alert.

Results

Diseases and symptoms dominate UpToDate searches. Some searches result in page views of only short duration, while others consistently result in longer-than-average page views. The response to an FDA alert for Celexa, characterized by a change in UpToDate search activity, differed considerably from general online media activity. Changes in search activity appeared later and persisted longer in UpToDate logs. The volume of searches and page view durations related to Celexa before the alert also differed from those after the alert.

Conclusions

Understanding the information-seeking behavior associated with online evidence sources can offer insight into the information needs of health professionals and enable large-scale medical surveillance. Our Web log mining approach has the potential to monitor responses to FDA alerts at a national level. Our findings can also inform the design and content of evidence-based medical information resources such as UpToDate.