Drug-induced comorbidities in patients with sarcoidosis.

PURPOSE OF REVIEW: Sarcoidosis is a chronic multisystemic inflammatory disease of unknown aetiology with a wide range of highly variable clinical manifestations and unpredictable disease course. Sarcoidosis patients may present with specific organ-related symptoms involving functional impairments, and less specific symptoms. The decision whether and when to treat a sarcoidosis patient with pharmacotherapy depends on two major factors: risk of organ failure and/or death and impairment of quality of life. This decision is complex and not standardized. RECENT FINDINGS: Glucocorticoids (GCs) are recommended as initial treatment, when needed. Subsequent GC-sparing alternatives frequently follow. Comorbidities or adverse drug reactions (ADRs) from drugs used in sarcoidosis treatment are sometimes very hard to differentiate from symptoms associated with the disease itself, which may cause diagnostic dilemmas. An ideal approach to minimalize ADRs would involve genetic screening prior to prescribing certain 'high-risk drugs' and therapeutic drug monitoring during treatment. Pharmacogenomic testing aims to guide appropriate selection of medicines, with the potential of reducing unnecessary polypharmacy while improving clinical outcomes. SUMMARY: A multidisciplinary approach to the management of sarcoidosis may avoid unnecessary ADRs. It is important to consider the possibility of drug-induced damage in sarcoidosis, especially if the clinical situation deteriorates after the introduction of a particular drug.

Body weight gain in clozapine-treated patients: Is norclozapine the culprit?

The antipsychotic drug clozapine is associated with weight gain. The proposed mechanisms include blocking of serotonin (5-HT2a/2c ), dopamine (D2 ) and histamine (H1 ) receptors. Clozapine is metabolized by cytochrome P450 1A2 (CYP1A2) to norclozapine, a metabolite with more 5-HT2c -receptor and less H1 blocking capacity. We hypothesized that norclozapine serum levels correlate with body mass index (BMI), waist circumference and other parameters of the metabolic syndrome. We performed a retrospective cross-sectional study in 39 patients (female n = 8 (20.5%), smokers n = 18 (46.2%), average age 45.8 ± 9.9 years) of a clozapine outpatient clinic in the Netherlands between 1 January 2017 and 1 July 2020. Norclozapine concentrations correlated with waist circumference (r = 0.354, P = .03) and hemoglobin A1c (HbA1c) (r = 0.34, P = .03). In smokers (smoking induces CYP1A2), norclozapine concentrations correlated with waist circumference (r = 0.723, P = .001), HbA1c (r = 0.49, P = .04) and BMI (r = 0.63, P = .004). Elucidating the relationship between norclozapine and adverse effects of clozapine use offers perspectives for interventions and treatment options.

Identifying adverse drug reactions from free-text electronic hospital health record notes.

BACKGROUND: Adverse drug reactions (ADRs) are estimated to be the fifth cause of hospital death. Up to 50% are potentially preventable and a significant number are recurrent (reADRs). Clinical decision support systems have been used to prevent reADRs using structured reporting concerning the patient's ADR experience, which in current clinical practice is poorly performed. Identifying ADRs directly from free text in electronic health records (EHRs) could circumvent this. AIM: To develop strategies to identify ADRs from free-text notes in electronic hospital health records. METHODS: In stage I, the EHRs of 10 patients were reviewed to establish strategies for identifying ADRs. In stage II, complete EHR histories of 45 patients were reviewed for ADRs and compared to the strategies programmed into a rule-based model. ADRs were classified using MedDRA and included in the study if the Naranjo causality score was ≥1. Seriousness was assessed using the European Medicine Agency's important medical event list. RESULTS: In stage I, two main search strategies were identified: keywords indicating an ADR and specific prepositions followed by medication names. In stage II, the EHRs contained a median of 7.4 (range 0.01-18) years of medical history covering over 35 000 notes. A total of 318 unique ADRs were identified of which 63 were potentially serious and 179 (sensitivity 57%) were identified by the rule. The method falsely identified 377 ADRs (positive predictive value 32%). However, it also identified an additional eight ADRs. CONCLUSION: Two key strategies were developed to identify ADRs from hospital EHRs using free-text notes. The results appear promising and warrant further study.

Inhibition of CYP2D6 with low dose (5 mg) paroxetine in patients with high 10-hydroxynortriptyline serum levels-A prospective pharmacokinetic study.

The antidepressant nortriptyline is metabolized by cytochrome P450 2D6 (CYP2D6) to the less active and more cardiotoxic drug metabolite, 10-hydroxynortriptyline. High serum levels of this metabolite (>200 µg/L) may lead to withdrawal of nortriptyline therapy. Adding CYP2D6 inhibitors reduce the metabolic activity of CYP2D6 (phenoconversion) and so decrease the forming of hydroxynortriptyline. In this study, 5 mg paroxetine is administered to patients with high hydroxynortriptyline concentrations (>200 µg/L). The shift in number of patients to therapeutic nortriptyline (50-150 µg/L) and safe hydroxynortriptyline (<200 µg/L) concentrations, and the degree of phenoconversion, expressed as the change in ratio nortriptyline/hydroxynortriptyline concentrations before and after paroxetine addition, are prospectively observed and described. After paroxetine addition, 12 patients (80%) had therapeutic nortriptyline and safe hydroxynortriptyline concentrations. Hydroxynortriptyline concentrations decreased in all patients. The average nortriptyline/hydroxynortriptyline concentrations ratio increased from 0.32 to 0.59. This study shows that 5 mg paroxetine addition is able to lower high hydroxynortriptyline serum levels to safe ranges.

Inflammatory bowel disease patients provide reliable self-reported medical information: A multicentre prospective pharmacovigilance monitoring system.

PURPOSE: To assess the agreement between patient-reported and health care provider-reported medical information in inflammatory bowel disease (IBD). METHODS: This multicentre, prospective, event monitoring study enrolled adult Crohn's disease (CD) and ulcerative colitis (UC) patients treated with a biological in four medical centers in the Netherlands. At two-monthly intervals, patients completed questionnaires on biological use, combination therapy and indication. The patient-reported information was compared with their electronic health records (EHRs) and analysed for percentage agreement and Cohen's kappa. A reference population from a prospective IBD registry was used to assess the representativeness of the study population. RESULTS: In total, 182 patients (female 50.5%, mean age 42.2 years, CD 76.9%) were included in the analysis. At baseline, 51.0% of the patients were prescribed an immunomodulator (43.9% thiopurines, 7.1% methotrexate), and patients were prescribed biologicals as follows: 59.3% infliximab, 30.2% adalimumab, 9.3% vedolizumab, and 1.1% ustekinumab. Agreement on patient-reported indication and biological use was almost perfect (κ = 0.878 and κ = 1.000, respectively); substantial for combination therapy (κ = 0.672). Gender, age, type of IBD, biological use and combination therapy were comparable with the reference population. CONCLUSION: Systematic patient-reporting by questionnaires was reliable in retrieving indication and treatment specific information from IBD patients. These results indicate that the use of patient-reporting outcomes in daily IBD practice can ensure reliable information collection.

Patients with inflammatory rheumatic diseases: quality of self-reported medical information in a prospective cohort event monitoring system.

OBJECTIVES: Assessment of the quality of patient-reported medical information in the Dutch Biologic Monitor and evaluation of the representativeness of the sampled participants. METHODS: Consecutive adult patients using a biologic DMARD (bDMARD) for an immune-mediated inflammatory disease were included in eight Dutch centres. For this substudy, data of 550 patients with inflammatory rheumatic diseases were used. Patient-reported bDMARD prescription, indication and combination therapy were verified for patients that permitted access to their electronic health record using percentage agreement and/or Cohen's kappa (n = 483). Conservative post hoc sensitivity analysis was performed to account for missing data. Population representativeness was tested for the entire substudy population by comparing age, gender and prescribed bDMARD to the centres' reference populations using Mann-Whitney U-test, χ2 goodness-of-fit or Fisher's exact test with Monte Carlo simulation (n = 550). RESULTS: The correct bDMARD was reported by 95.8% of the participants. Agreement between patients and electronic health record was almost perfect for indications (κ = 0.832) and substantial for combination therapies (κ = 0.725). Agreement on combination therapies remained substantial after post hoc sensitivity analysis (κ = 0.640). Gender distribution (P > 0.05) and bDMARD use (P > 0.05) were similar to the reference populations. Median age was different (58.0 vs 56.0 years, P = 0.04), but considered clinically irrelevant. CONCLUSION: The Dutch Biologic Monitor seems to be a valid tool to obtain patient-reported medical information. Reported medical information generally corresponded to the electronic health records and the participants represented their reference populations regarding age, gender and prescribed bDMARD.

Tamsulosin Associated with Interstitial Lung Damage in CYP2D6 Variant Alleles Carriers.

Drugs are serious but underestimated causative agents of interstitial lung disease (ILD). Both cytotoxic and immune mechanisms may be involved in drug-induced ILD (DI-ILD). We aimed to investigate whether polymorphisms of relevant CYP enzymes involved in the metabolization of tamsulosin might explain the pathologic mechanism of the DI-ILD in the cases with suspected tamsulosin DI-ILD. We collected 22 tamsulosin-associated DI-ILD cases at two ILD Expertise Centers in the Netherlands between 2009 and 2020. CYP2D6, CYP2C9, CYP2C19, CYP3A4, and CYP3A5 single nucleotide polymorphisms were genotyped and compared with a control group of 78 healthy Caucasian male volunteers. Nine cases were phenotyped as CYP2D6 poor metabolizers and 13 as CYP2D6 intermediate metabolizers. The phenotypes of the cases differed significantly from those of the healthy controls, with more poor metabolizers. After withdrawal of tamsulosin, the pulmonary condition of three cases had improved, six patients had stabilized, and one patient stabilized after reducing the tamsulosin dose. The described 22 cases suggest that an association between the presence of CYP2D6 allelic variants and tamsulosin-associated ILD is highly likely. These cases highlight the importance of both clinical and genetic risk stratification aimed to achieve a more accurate prevention of DI-ILD in the future and enhance the quality of life of patients.

Drug-induced interstitial lung disease: role of pharmacogenetics in predicting cytotoxic mechanisms and risks of side effects.

PURPOSE OF REVIEW: The diagnosis of drug-induced interstitial lung disease (DI-ILD) is challenging and mainly made by exclusion of other possible causes. Toxicity can occur as a cause of drug(s) or drug-drug interactions. In this review, we summarize the possible role of pharmacogenetics of metabolizing enzymes in DI-ILD. RECENT FINDINGS: Knowledge of the genetic predispositions of enzymes involved in drug metabolization and their relation with proposed cytotoxic mechanisms of DI-ILD, in particular direct cell toxicity and free oxygen radical production is increasing. The cytochrome P450 enzyme family and other enzymes play an important role in the metabolism of all sorts of ingested, injected, or inhaled xenobiotic substances. The liver is the major site for metabolism. Metabolic cytotoxic mechanisms have however also been detected in lung tissue. Polymorphisms in genes coding for enzymes that influence metabolic activity may lead to localized (toxic) reactions and tissue damage. This knowledge may be helpful in preventing the risk of DI-ILD. SUMMARY: Drug toxicity can be the consequence of absence or very poor enzyme activity, especially if no other metabolic route is available. In the case of reduced enzyme activity, it is recommended to reduce the dose or to prescribe an alternative drug, which is metabolized by a different, unaffected enzyme system to prevent toxic side effects. However, enhanced enzyme activity may lead to excessive formation of toxic and sometimes reactive metabolites. Therefore, knowing a patient's drug-metabolizing profile before drug prescription is a promising way to prevent or explain DI-ILD.

Monitoring Haloperidol Plasma Concentration and Associated Adverse Events in Critically Ill Children With Delirium: First Results of a Clinical Protocol Aimed to Monitor Efficacy and Safety.

OBJECTIVES: As delirium in critically ill children is increasingly recognized, more children are treated with the antipsychotic drug haloperidol, while current dosing guidelines are lacking solid evidence and appear to be associated with a high risk of adverse events. We aim to report on the safety and efficacy of a recently implemented clinical dose-titration protocol with active monitoring of adverse events. DESIGN: From July 2014 until June 2015, when a potential delirium was identified by regular delirium scores and confirmed by a child psychiatrist, haloperidol was prescribed according to the Dutch Pediatric Formulary. Daily, adverse events were systematically assessed, haloperidol plasma concentrations were measured, and delirium symptoms followed. Dependent on the clinical response, plasma concentration, and adverse event, the dose was adjusted. SETTING: A 28-bed tertiary PICU in the Netherlands. PATIENTS: All patients admitted to the PICU diagnosed with delirium. INTERVENTION: Treatment with haloperidol according to a dose-titration protocol MEASUREMENTS AND MAIN RESULTS:: Thirteen children (median age [range] 8.3 yr [0.4-13.8 yr]) received haloperidol, predominantly IV (median dose [range] 0.027 mg/kg/d [0.005-0.085 mg/kg/d]). In all patients, pediatric delirium resolved, but five of 13 patients developed possible adverse event. These were reversed after biperiden (n = 2), discontinuing (n = 3), and/or lowering the dose (n = 3). Plasma concentrations were all below the presumed therapeutic threshold of 3-12 µg/L. CONCLUSIONS: Prospective systematic monitoring of adverse event in critically ill children receiving haloperidol revealed a significant proportion of possible adverse events. Adverse event developed despite low plasma concentrations and recommended dose administration in the majority of the patients. Our data suggest that haloperidol can potentially improve pediatric delirium, but it might also put patients at risk for developing adverse events.

Burden of non-serious infections during biological use for rheumatoid arthritis.

INTRODUCTION: Biologicals have become a cornerstone in rheumatoid arthritis (RA) treatment. The increased risk of serious infections associated with their use is well-established. Non-serious infections, however, occur more frequently and are associated with a high socioeconomic burden and impact on quality of life but have not received the same attention in the literature to date. The aim of this study was to gain insight into the various non-serious infections reported in RA patients using biologicals and their experienced burden. MATERIALS AND METHODS: The Dutch Biologic Monitor was a prospective observational study that included adults with rheumatoid arthritis and biological use who answered bimonthly questionnaires on the adverse drug reactions (ADRs) they experienced from their biological and reported the associated impact score (ranging from 1, no impact, to 5, very high impact). ADRs were assigned a MedDRA code by pharmacovigilance experts and labeled as definite, probable, possible or no infection by infectious disease professionals. Descriptive statistics were performed using medians and interquartile ranges. RESULTS: A total of 586 patients were included in the final analysis. Eighty-five patients (14.5%) reported a total of 421 ADRs labeled as probable or definite infections by the experts. Patient-assigned burden was ADR-specific. Upper respiratory tract infections were most frequently reported and had a high rate of recurrence or persistence, with a median impact score of 3.0 (IQR 2.0-3.0) which remained stable over time. DISCUSSION: Non-serious infections significantly outnumbered serious infections in this real-life cohort of RA patients using biologicals (77.1 non-serious infections and 1.3 serious infections per 100 patient years, respectively). Infections in the upper respiratory tract were rated as having an average burden, which remained constant over a long period of time. Awareness of the impact of recurrent and chronic non-serious infections may enable healthcare professionals to timely treat and maybe even prevent them, which would lessen the associated personal and socioeconomic burden.

Drug-Gene Risk Stratification in Patients with Suspected Drug-Induced Interstitial Lung Disease.

BACKGROUND: Pulmonary toxicity has been associated with drug use. This is often not recognized in clinical practice, and underestimated. OBJECTIVE: We aimed to establish whether polymorphisms in certain genes corresponding with a metabolic pathway of drug(s) used are associated with pulmonary toxicity in patients with suspected drug-induced interstitial lung disease (DI-ILD). METHODS: This retrospective observational study explored genetic variations in three clinically relevant cytochrome P450 (CYP) iso-enzymes (i.e., CYP2D6, CYP2C9, and CYP2C19) in a group of patients with a fibroticinterstitial lung disease, either non-specific interstitial pneumonia (n = 211) or idiopathic pulmonary fibrosis (n = 256), with a suspected drug-induced origin. RESULTS: Of the 467 patients, 79.0% showed one or more polymorphisms in the tested genes accompanied by the use of drug(s) metabolized by a corresponding affected metabolic pathway (60.0% poor metabolizers and/or using two or more drugs [likely DI-ILD], 37.5% using three or more [highly likely DI-ILD]). Most commonly used drugs were statins (63.1%) with a predominance among men (69.4 vs 47.1%, p < 0.0001). Nitrofurantoin, not metabolized by the tested pathways, was prescribed more frequently among women (51.9 vs 4.5%, p < 0.00001). CONCLUSIONS: In our cohort with suspected DI-ILD, 79% carried one or more genetic variants accompanied by the use of drugs metabolized by a corresponding affected pathway. In 60%, the diagnosis of DI-ILD was likely, whereas in 37.5%, it was highly likely, based on CYP analyses. This study underlines the importance of considering both drug use and genetic make-up as a possible cause, or at least a contributing factor, in the development and/or progression of fibrotic lung diseases. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov identifier NCT00267800, registered in 2005.

The burden of adverse drug reactions reported by patients in the Dutch ADR monitor: a proof of concept.

BACKGROUND: We aimed to describe the burden of adverse drug reactions (ADRs) reported by patients participating in the Dutch ADR Monitor using a multifactorial burden measurement instrument. METHODS: The Dutch ADR Monitor is a cohort event monitoring system that collects information on ADR experiences, including burden. This study includes the initial data (November 2022 until May 2023). Patients were asked if experienced ADRs impacted 7 domains of burden: appearance, medical treatment, daily life, fatigue, physical consequences, mental consequences and the course of ADRs. Burden was scored from 0 to 10 on impacted domains. The distributions of these burden scores were demonstrated in Likert plots. The burden between persistent and recurrent ADRs was compared. RESULTS: 92 patients reported 199 ADRs. Impact on the domains fatigue and daily life were experienced most frequently, except for skin and subcutaneous tissue ADRs, where impact on appearance and mental consequences were experienced most frequently. Fatigue was considered the most burdensome domain. No difference in burden was found between persistent (median = 7, IQR = 4) and recurrent ADRs (median = 6, IQR = 4, p = 0.59). CONCLUSIONS: This is the first study investigating burden of ADRs on 7 domains in patients with chronic diseases. Impact on the domain fatigue was considered most burdensome.

Patients with skin and subcutaneous ADRs experienced impact on appearance and mental consequences most often, but found impact on fatigue most burdensome.For most reported ADRs, patients scored the highest burden on the domain fatigue and thus found impact on this domain to be the most burdensome.Patients with skin and subcutaneous ADRs experienced impact on appearance and mental consequences most often, but found impact on fatigue most burdensome.

Development of a text mining algorithm for identifying adverse drug reactions in electronic health records.

Objective: Adverse drug reactions (ADRs) are a significant healthcare concern. They are often documented as free text in electronic health records (EHRs), making them challenging to use in clinical decision support systems (CDSS). The study aimed to develop a text mining algorithm to identify ADRs in free text of Dutch EHRs. Materials and Methods: In Phase I, our previously developed CDSS algorithm was recoded and improved upon with the same relatively large dataset of 35 000 notes (Step A), using R to identify possible ADRs with Medical Dictionary for Regulatory Activities (MedDRA) terms and the related Systematized Nomenclature of Medicine Clinical Terms (SNOMED-CT) (Step B). In Phase II, 6 existing text-mining R-scripts were used to detect and present unique ADRs, and positive predictive value (PPV) and sensitivity were observed. Results: In Phase IA, the recoded algorithm performed better than the previously developed CDSS algorithm, resulting in a PPV of 13% and a sensitivity of 93%. For The sensitivity for serious ADRs was 95%. The algorithm identified 58 additional possible ADRs. In Phase IB, the algorithm achieved a PPV of 10%, a sensitivity of 86%, and an F-measure of 0.18. In Phase II, four R-scripts enhanced the sensitivity and PPV of the algorithm, resulting in a PPV of 70%, a sensitivity of 73%, an F-measure of 0.71, and a 63% sensitivity for serious ADRs. Discussion and Conclusion: The recoded Dutch algorithm effectively identifies ADRs from free-text Dutch EHRs using R-scripts and MedDRA/SNOMED-CT. The study details its limitations, highlighting the algorithm's potential and significant improvements.

Quantification of longitudinal patient-reported burden of adverse drug reactions attributed to the use of TNF-α inhibitors in inflammatory rheumatic diseases: an observational prospective cohort study.

BACKGROUND: There is a lack of knowledge on patient perspectives on adverse drug reactions (ADRs) attributed to the use of biologics. The aim of this study is to quantify the burden over time of ADRs attributed to TNF-α inhibitors in patients with inflammatory rheumatic diseases (IRDs) and investigate whether the burden over time differs between different types of ADRs. RESEARCH DESIGN AND METHODS: Data were used from the Dutch Biologic Monitor (DBM), an observational prospective cohort study for patient-reported ADRs attributed to biologics. Patients with an IRD using a TNF-α inhibitor reporting an ADR, lasting for three consecutive questionnaires, were included. Questionnaires were sent every 2 months and the burden was scored on a 5-point Likert-type scale. Burden scores were analyzed using linear mixed models. RESULTS: Data from 166 unique patients reporting 274 ADRs were included. The burden score decreased every month by 0.29 points (95% CI -0.34 - -0.24) on average on a 5-point Likert-type scale. The burden score for infections and infestations decreased significantly faster than the burden score for injection site reactions. CONCLUSIONS: Patient-reported burden of ADRs attributed to the use of a TNF-α inhibitor in patients with IRDs decreased significantly over time, especially for infections and infestations.

The effect of corticosteroids, antibiotics, and anticoagulants on the development of post-COVID-19 syndrome in COVID-19 hospitalized patients 6 months after discharge: a retrospective follow up study.

To assess the effect of pharmacotherapeutic interventions commonly employed in the management of COVID-19 hospitalized patients on the development of post-COVID-19 syndrome. This study employed two distinct databases, the Medisch Spectrum Twente (MST) clinical database comprising electronic health records of COVID-19 patients hospitalized at MST, and the Post-COVID cohort database which contains follow-up information on the same patients. These databases were integrated to establish the potential relationship between the administration of corticosteroids, antibiotics, or anticoagulants during hospitalization and the occurrence of post-COVID-19 syndrome after a 6-month interval following discharge. A total of 123 patients who were hospitalized due to COVID-19 infection were included in this study. Among these patients, 33 (26.8%) developed post-COVID-19 syndrome which persisted even 6 months after hospital discharge. Multivariate analysis revealed that patients who received treatment with corticosteroids had a significantly lower likelihood (OR 0.32, 95% CI 0.11-0.90) of developing post-COVID-19 syndrome, while no significant association was observed for treatment with antibiotics (OR 1.26, 95% CI 0.47-3.39) or anticoagulants (OR 0.55, 95% CI 0.18-1.71). The findings of this study indicate that corticosteroids exert a significant protective effect against the development of post-COVID-19 syndrome in patients who were hospitalized due to COVID-19 infection. Although a trend towards a protective effect of anticoagulants was observed, it did not reach statistical significance. On the contrary, patients treated with antibiotics were shown to have increased chances of developing post-COVID-19 syndrome, although this effect was also not statistically significant.

Impact of adverse drug reactions on the treatment pathways of early rheumatoid arthritis patients: a prospective observational cohort study.

BACKGROUND: Several patient characteristics may be of influence on treatment pathways of rheumatoid arthritis (RA) patients in clinical practice. The aim of this study is to analyze treatment pathways of early RA patients stratified for gender and adverse drug reaction (ADR) occurrence. RESEARCH DESIGN AND METHODS: Treatment pathways of patients included in the DREAM-RA treat-to-target cohort I between 16th of July 2006-30th of April 2020 were assessed. Treatment pathways were visualized in Sankey diagrams. Follow-up time, duration per treatment and the number of treatments received were stratified for gender and ADR occurrence and analyzed. Independent t-tests and chi-square tests were performed where applicable. RESULTS: Treatment pathways of 372 patients (follow-up: 2488.4 years, mean 6.7 ± 3.7 years) were analyzed. The Sankey diagrams visualize that treatment pathways became increasingly varied and complex over time. No significant differences were found when comparing female patients and male patients. However, the average treatment duration was shorter in patients with ADRs (1.8 vs. 2.7 years, p < 0.05), and the number of treatments higher (3.5 vs. 2.5, p < 0.05). CONCLUSIONS: Treatment pathways increase in complexity over time. Differences were found between patients with and without ADRs, with patients that experience ADRs receiving more and shorter treatments.

Development of a Framework Structuring Themes in the Course of Adverse Drug Reactions from a Patient's Perspective.

INTRODUCTION: There is a need for more extensive information about adverse drug reactions (ADRs) for patients than currently available, including information on the course of ADRs. Aspects characterising the course of ADRs from the patient perspective have not been identified before. OBJECTIVE: We aimed to develop a framework based on common themes in the course of ADRs identified from patient descriptions in patient-reported ADRs. METHODS: In this qualitative study, patient descriptions of the course of patient-reported ADRs were analysed by a thematic analysis with an inductive approach using three different existing datasets containing patient-reported ADRs. Two datasets included patient-reported ADRs from cohort event monitoring of biologics and direct oral anticoagulants and one dataset included spontaneous reports from patients concerning medication for lower urinary tract symptoms. A conceptual framework was developed from the identified main themes and subthemes. RESULTS: Patient-reported data concerning 3888 ADRs were analysed. Six main themes with multiple subthemes were identified from patient descriptions of the course of ADRs. Four themes were descriptive: frequency of an ADR episode, duration of an ADR episode, moment or period of ADR occurrence, and development in the intensity of the ADR. Two themes concerned factors influencing the course of ADRs: triggering factors and improving factors. CONCLUSIONS: The presented framework illustrates that patients describe extensive details on the course and timeframe of ADRs. The identified themes provide a basis for improving the systematic data collection of more extensive details about ADRs from patients as a first step towards the provision of more comprehensive ADR information to patients.