Real-World Evidence on Levetiracetam-Induced Hypokalemia: An Active Comparator Cohort Study.

BACKGROUND: Levetiracetam is an anti-seizure medication (ASM) with an established safety profile. However, a potential safety signal of hypokalemia following levetiracetam use was published in the World Health Organization newsletter. OBJECTIVE: To investigate the possible causal association between the use of levetiracetam and the development of hypokalemia. METHOD: This was a new-user, active-comparator retrospective cohort study using Real-world Evidence Research Network data at the Saudi Food and Drug Authority from 2016 to 2022. Adults (≥ 18 years old) with an incident prescription for either levetiracetam or carbamazepine were followed for up to 6 months from the prescription date. Hypokalemia was ascertained by using diagnostic code (i.e., E87.6) or by serum potassium level below 3.5 mmol/L. A Cox proportional hazards model, adjusted with stabilized inverse probability of treatment weight, was fitted to compare the hazard of hypokalemia between levetiracetam and carbamazepine exposed patients. RESULTS: A total of 8,982 patients entered the study cohort. The incidence rate of hypokalemia was 303 cases per 10,000 patient-years in the levetiracetam-exposed cohort compared to 57 cases per 10,000 patient-years among carbamazepine users. Compared to carbamazepine users, patients exposed to levetiracetam had an adjusted hazard ratio related to induced hypokalemia of 1.99 (95% confidence interval, 0.88-4.49). Results of sensitivity analyses were comparable to the main analysis. CONCLUSION: The hazard ratio for hypokalemia with the use of levetiracetam versus carbamazepine was statistically comparable. However, the potential association between levetiracetam use and hypokalemia cannot be ruled out given the elevated hazard ratios from the main and sensitivity analyses. Further studies may provide a more precise assessment of this association.

The inadvertent consequences of drug recalls: A case study of a recall of pantoprazole generics from the markets.

Introduction: Drug recalls may impact treatment plans or access to suitable therapies. Thus, they inadvertently affect treatment outcomes. Objective: We aimed to examine the impact of recalls on patients' safety using pantoprazole-containing products recall as a case study in terms of the occurrence of potential drug-drug interactions (pDDIs). Methods: This retrospective study used de-identified electronic health records of adult patients who had a prescription for oral proton pump inhibitors (PPIs) including pantoprazole, esomeprazole, lansoprazole, or omeprazole from April 2020 through September 2021 from a large tertiary care hospital. The study outcome definition was the prevalence of pDDIs in PPIs users before and after the recall date (March 2021). Changes in the prevalence of pDDIs were modeled using interrupted time-series. The rate ratio of pDDIs in the 12 months before and 6 months after the recall was modeled using negative binomial regression. Results: A total of 1,826 pDDIs were identified, and the median monthly prevalence of pDDI before the recall was 102.5 which increased to 115.5 after the recall. A change in the level of pDDIs occurred immediately after the recall date, followed by a gradual decrease over time. The rate of pDDIs was 69% higher after the recall compared to the baseline (rate ratio 1.69; 95% confidence interval, 0.75-1.91). Discussion: Recall of pantoprazole-containing products was associated with a higher rate of pDDIs. However, the prevalence of pDDIs gradually decreased over time. We highlight the importance of planning of recall process and coordinating all potential stakeholders to avoid potential harms.Word count: 1450.

Real-world data in Saudi Arabia: Current situation and challenges for regulatory decision-making.

PURPOSE: To present the process of establishing a pharmacoepidemiological database in Saudi Arabia, challenges and models used. METHODS: The database establishment has started in 2017 by piloting the conversion of electronic health records of one hospital to the Observational Health Data Sciences and Informatics (OHDSI), Observational Medical Outcomes Partnership's Common Data Model (OMOP). RESULTS: During the pilot phase we have faced several challenges such as limited contribution in providing data by local medical institution due to uncertainty about data governance, diversity of systems used by hospitals, inconsistent coding of medical information, and limited awareness about data structure from participating hospital. The pilot phase was completed in 2019 containing information about patient attributes, medical care, therapies, and other additional services for around 130 000 patients in Saudi Arabia. The majority of patients were below the age of 50 years (89%), and acute respiratory infections were the most frequent diagnosis. The data quality was acceptable and no major anomalies were detected during the conversion. CONCLUSIONS: We demonstrated a successful creation of a pilot database using OHDSI Common Data Model. Our experience with the pilot database could be extended to other institutions to create a national dataset that could be used to generate real-world evidence.

Concomitant Use of Quinolones and Stimulants and the Risk of Adverse Cardiovascular Symptoms: A Retrospective Cohort Study.

OBJECTIVE: To examine whether concomitant use of quinolones and stimulants increases the risk of cardiac events in adults. STUDY DESIGN: A retrospective cohort study of privately insured adults using MarketScan® claims data from 2008 to 2015. PATIENTS: Stimulant (methylphenidate or mixed amphetamine salts) users (18-65 yrs old) with continuous health plan enrollment for the 6 months (baseline) prior to the first dispensation (index date) of oral quinolones or comparators (amoxicillin ± clavulanate or azithromycin). OUTCOMES DEFINITION: (1) Cardiac symptoms (palpitation, tachycardia, or syncope); (2) cardiac arrhythmias (ventricular arrhythmias, paroxysmal ventricular tachycardia, or cardiac arrest). ANALYSIS: Baseline covariates adjustment was through inverse probability of treatment weighting. Adults were followed until the antimicrobial therapy ended. The hazard of cardiac events in stimulant-quinolones-exposed adults was compared to those who were treated with stimulant-comparator antibiotics using a weighted Cox regression model. Several sensitivity analyses were performed to challenge the results robustness. RESULTS: The study cohorts comprised 390,490 stimulants users who initiated either quinolone or amoxicillin, and 387,574 patients receiving stimulants who initiated quinolone or azithromycin. The unadjusted incidence rate for cardiac symptoms in stimulant-quinolones users was 471 cases/10,000 patient-years, and it was 244 cases/10,000 patient-years in patients exposed to stimulant-amoxicillin; whereas the unadjusted incidence rate for cardiac symptoms was 728 and 358 per 10,000 patient-years for stimulant-quinolones and stimulant-azithromycin cohorts, respectively. Compared to stimulant-amoxicillin use, the adjusted hazard ratio (HR) for cardiac symptoms with stimulant-quinolones use was 1.61 (95% confidence interval [CI], 1.30-1.98). The HR for cardiac symptoms for patient exposed to stimulant-quinolones was 1.69 (95% CI, 1.32-2.13) when compared to stimulant-azithromycin. The sensitivity analysis findings were consistent with the primary analysis. A few patients across the study comparison groups developed cardiac arrhythmias. CONCLUSION: Concomitant use of stimulants and quinolone was associated with an increased hazard of cardiac symptoms in comparison to concomitant use of stimulants and amoxicillin or azithromycin, but there was no apparent difference in cardiac arrhythmias.

Model-based standardization using an outcome model with random effects.

Model-based standardization uses a statistical outcome model or exposure model to estimate a population-average association that is unconfounded by selected covariates. With it, one can compare groups using a distribution of confounders identical in each group to that of a standard population. We develop an approach based on an outcome model, in which the mean of the outcome is modeled conditional on the exposure and the confounders. In our approach, there is a confounder that clusters the observations into a very large number of categories. We treat the parameters for the clusters as random effects. We use a between-within model to account for the association of the random effects not only with the exposure but also with the cluster population sizes. We review alternative approaches presented in the literature, and we compare the outcome-modeling approach to recently proposed exposure-modeling approaches incorporating random effects. To illustrate, we use 2014 to compare proportions of acute respiratory tract infection diagnoses with an antibiotic prescription for emergency department versus outpatient visits, adjusting for confounding by unmeasured patient level variables and measured diagnosis-level variables. We also present results of a simulation study.

Trends of Neonatal Abstinence Syndrome Epidemic and Maternal Risk Factors in Florida.

OBJECTIVE: This study aimed to examine secular trends of (i) maternal prescription opioid use in late pregnancy, (ii) neonatal abstinence syndrome (NAS) stratified by late maternal prescription opioid use, and (iii) maternal risk factors among NAS deliveries. METHODS: Women with a live birth who were enrolled 90 days before and 30 days after delivery in Florida Medicaid Analytic Extract billing records linked to birth certificates from 2000 to 2010 were identified for the study. Changes in the annual prevalence of prescription opioid use during pregnancy were tested with use of the Cochran-Armitage trend test. Temporal trends of NAS deliveries were estimated with the use of Poisson regression and stratified by prescription opioid exposure in the last 90 days of pregnancy in the study period. To identify contributors to the increase in NAS cases, variations in prevalence of opioid dispensing, tobacco use, antidepressant use, and substance use disorder among NAS and non-NAS deliveries were examined. RESULTS: There were 41,968 (9.4%) deliveries exposed to at least one opioid prescription in late pregnancy, and this rate remained stable from 2000 to 2010. Among prescription opioid-exposed deliveries, frequency of NAS increased from 1.6 to 25.2 per 1000 live births during the study period (p<0.05). Although the prevalence of maternal use of prescription opioid, tobacco, and antidepressants remained stable among NAS deliveries from 2000 to 2010, the prevalence of substance use disorder diagnoses increased substantially from 38.9% in 2000 to 67.9% in 2006 (p<0.05). CONCLUSIONS: The prevalence of NAS increased dramatically whereas the prevalence of major risk factors, including maternal prescription opioid use, remained stable in Florida between 2000 and 2010. The increase in substance use disorder may be responsible for the sharp increase in NAS deliveries.

Knowledge and attitude of healthcare professionals toward medication pregnancy category systems in Saudi hospitals.

OBJECTIVE: To assess the knowledge and attitudes of healthcare professionals (HCPs) toward systems used in describing the safety of medications use during pregnancy. METHODS: A cross-sectional self-administered survey was conducted in 4 tertiary hospitals in Riyadh, Saudi Arabia between March and May 2012. The targeted HCPs were physicians and pharmacists. The survey was validated and contained 4 main sections. Descriptive statistics were used to report responses to the survey's questions. RESULTS: A total of 393 HCPs responded to the survey, with a response rate of 97%. Half of the respondents were physicians. Of the participants, 60% were males. Most respondents (66%) stated that they have prescribed/dispensed a drug that may cause teratogenicity. Moreover, 87% of the respondents (48% pharmacists and 39% physicians) were aware of the Food and Drug Administration (FDA) pregnancy category, and most (72%) found it helpful. Only 11% of the participants strongly agree to use the European Medicine Agency (EMA) system for pregnancy category system as their main reference. CONCLUSION: In general, HCPs in Saudi Arabian hospitals have good knowledge of and attitudes toward pregnancy category systems, with more familiarity with the FDA system. The FDA system is preferred over the EMA system.

Community pharmacists' knowledge of isotretinoin safety.

OBJECTIVE: To explore the practice and knowledge of community pharmacists in Saudi Arabia regarding dispensing isotretinoin-containing products. METHODS: This was a cross-sectional study conducted in 2012 that included community pharmacists from 3 cities in Saudi Arabia. A validated and piloted self-administered survey collected demographics and information on the pharmacist's knowledge regarding isotretinoin precautions, as well as his dispensing, and counseling practices. RESULTS: One hundred and sixteen questionnaires were returned with a 72.5% response rate. Only around half of the participants (56%) knew the correct pregnancy risk classification category for oral isotretinoin. Most participants (78%) correctly identified teratogenicity as the most serious risk associated with the use of oral isotretinoin. However, only 6.2% of the pharmacists recommended using 2 methods of contraception. Almost one-fifth of the pharmacists dispensed isotretinoin without a prescription. Finally, 11% of the pharmacists did not ask whether the patient performed a pregnancy test prior to dispensing oral isotretinoin. CONCLUSION: Pharmacists at community pharmacies are not adequately aware of the risks for female patients using isotretinoin. Additionally, an alarming proportion of pharmacists dispense isotretinoin without a prescription. It is essential to implement risk minimization plans for certain medications to limit and prevent adverse drug reactions.