Treatment complexity rises in line with the number of drugs, single doses, and administration methods, thereby threatening patient adherence. Patients with multimorbidity often need flexible, individualised treatment regimens, but alterations during the course of treatment may further increase complexity. The objective of our study was to explore medication changes in older patients with multimorbidity and polypharmacy in general practice. We retrospectively analysed data from the cluster-randomised PRIMUM trial (PRIoritisation of MUltimedication in Multimorbidity) conducted in 72 general practices. We developed an algorithm for active pharmaceutical ingredients (API), strength, dosage, and administration method to assess changes in physician-reported medication data during two intervals (baseline to six-months: ∆1; six- to nine-months: ∆2), analysed them descriptively at prescription and patient levels, and checked for intervention effects. Of 502 patients (median age 72 years, 52% female), 464 completed the study. Changes occurred in 98.6% of patients (changes were 19% more likely in the intervention group): API changes during ∆1 and ∆2 occurred in 414 (82.5%) and 338 (67.3%) of patients, dosage alterations in 372 (74.1%) and 296 (59.2%), and changes in API strength in 158 (31.5%) and 138 (27.5%) respectively. Administration method changed in 79 (16%) of patients in both ∆1 and ∆2. Simvastatin, metformin and aspirin were most frequently subject to alterations. Medication regimens in older patients with multimorbidity and polypharmacy changed frequently. These are mostly due to discontinuations and dosage alterations, followed by additions and restarts. These findings cast doubt on the effectiveness of cross-sectional assessments of medication and support longitudinal assessments where possible. 1. Prospective registration: Trial registration number: NCT01171339 ; Name of registry: ClinicalTrials.gov; Date of registration: July 27, 2010; Date of enrolment of the first participant to the trial: August 12, 2010. 2. Peer reviewed trial registration: Trial registration number: ISRCTN99526053 ; Name of registry: Controlled Trials; Date of registration: August 31, 2010; Date of enrolment of the first participant to the trial: August 12, 2010.

Hip fractures are among the most frequently occurring fragility fractures in older adults, associated with a loss of quality of life, high mortality, and high use of healthcare resources. The aim was to apply the superlearner method to predict osteoporotic hip fractures using administrative claims data and to compare its performance to established methods. We devided claims data of 288,086 individuals aged 65 years and older without care level into a training (80%) and a validation set (20%). Subsequently, we trained a superlearner algorithm that considered both regression and machine learning algorithms (e.g., support vector machines, RUSBoost) on a large set of clinical risk factors. Mean squared error and measures of discrimination and calibration were employed to assess prediction performance. All algorithms used in the analysis showed similar performance with an AUC ranging from 0.66 to 0.72 in the training and 0.65 to 0.70 in the validation set. Superlearner showed good discrimination in the training set but poorer discrimination and calibration in the validation set. The superlearner achieved similar predictive performance compared to the individual algorithms included. Nevertheless, in the presence of non-linearity and complex interactions, this method might be a flexible alternative to be considered for risk prediction in large datasets.

Cardiovascular disease is a leading cause of death in older people, and the impact of being exposed or not exposed to preventive cardiovascular medicines is accordingly high. Underutilization of beneficial drugs is common, but prevalence estimates differ across settings, knowledge on predictors is limited, and clinical consequences are rarely investigated. Using data from a prospective population-based cohort study, we assessed the prevalence, determinants, and outcomes of medication underuse based on cardiovascular criteria from Screening Tool To Alert to Right Treatment (START). Medication underuse was present in 69.1% of 1454 included participants (mean age 71.1 ± 6.1 years) and was significantly associated with frailty (odds ratio: 2.11 [95% confidence interval: 1.24–3.63]), body mass index (1.03 [1.01–1.07] per kg/m²), and inversely with the number of prescribed drugs (0.84 [0.79–0.88] per drug). Using this information for adjustment in a follow-up evaluation (mean follow-up time 2.24 years) on cardiovascular and competing outcomes, we found no association of medication underuse with cardiovascular events (fatal and non-fatal) (hazard ratio: 1.00 [0.65–1.56]), but observed a significant association of medication underuse with competing deaths from non-cardiovascular causes (2.52 [1.01–6.30]). Medication underuse was associated with frailty and adverse non-cardiovascular clinical outcomes. This may suggest that cardiovascular drugs were withheld because of serious co-morbidity or that concurrent illness can preclude benefit from cardiovascular prevention. In the latter case, adapted prescribing criteria should be developed and evaluated in those patients.

Purpose: We assessed the differential effect of clarithromycin, a strong inhibitor of cytochrome P450 (CYP) 3A4 and P-glycoprotein, on the pharmacokinetics of a regular dose of edoxaban and on a microdose cocktail of factor Xa inhibitors (FXaI). Concurrently, CYP3A activity was determined with a midazolam microdose. Methods: In an open-label fixed-sequence trial in 12 healthy volunteers, the pharmacokinetics of a microdosed FXaI cocktail (μ-FXaI; 25 μg apixaban, 50 μg edoxaban, and 25 μg rivaroxaban) and of 60 mg edoxaban before and during clarithromycin (2 x 500 mg/d) dosed to steady-state was evaluated. Plasma concentrations of study drugs were quantified using validated ultra-performance liquid chromatography–tandem mass spectrometry methods. Results: Therapeutic clarithromycin doses increased the exposure of a therapeutic 60 mg dose of edoxaban with a geometric mean ratio (GMR) of the area under the plasma concentration-time curve (AUC) of 1.53 (90 % CI: 1.37–1.70; p < 0.0001). Clarithromycin also increased the GMR (90% CI) of the exposure of microdosed FXaI apixaban to 1.38 (1.26–1.51), edoxaban to 2.03 (1.84–2.24), and rivaroxaban to 1.44 (1.27–1.63). AUC changes observed for the therapeutic edoxaban dose were significantly smaller than those observed with the microdose (p < 0.001). Conclusion: Clarithromycin increases FXaI exposure. However, the magnitude of this drug interaction is not expected to be clinically relevant. The edoxaban microdose overestimates the extent of the drug interaction with the therapeutic dose, whereas AUC ratios for apixaban and rivaroxaban were comparable to the interaction with therapeutic doses as reported in the literature. Trial Registration: EudraCT Number: 2018-002490-22

Background: Decision making for the “best” treatment is particularly challenging in situations in which individual patient response to drugs can largely differ from average treatment effects. By estimating individual treatment effects (ITEs), we aimed to demonstrate how strokes, major bleeding events, and a composite of both could be reduced by model-assisted recommendations for a particular direct oral anticoagulant (DOAC). Methods: In German claims data for the calendar years 2014–2018, we selected 29 901 new users of the DOACs rivaroxaban and apixaban. Random forests considered binary events within 1 y to estimate ITEs under each DOAC according to the X-learner algorithm with 29 potential effect modifiers; treatment recommendations were based on these estimated ITEs. Model performance was evaluated by the c-for-benefit statistics, absolute risk reduction (ARR), and absolute risk difference (ARD) by trial emulation. Results: A significant proportion of patients would be recommended a different treatment option than they actually received. The stroke model significantly discriminated patients for higher benefit and thus indicated improved decisions by reduced outcomes (c-for-benefit: 0.56; 95% confidence interval [0.52; 0.60]). In the group with apixaban recommendation, the model also improved the composite endpoint (ARR: 1.69 % [0.39; 2.97]). In trial emulations, model-assisted recommendations significantly reduced the composite event rate (ARD: −0.78 % [−1.40; −0.03]). Conclusions: If prescribers are undecided about the potential benefits of different treatment options, ITEs can support decision making, especially if evidence is inconclusive, risk-benefit profiles of therapeutic alternatives differ significantly, and the patients’ complexity deviates from “typical” study populations. In the exemplary case for DOACs and potentially in other situations, the significant impact could also become practically relevant if recommendations were available in an automated way as part of decision making. Highlights It was possible to calculate individual treatment effects (ITEs) from routine claims data for rivaroxaban and apixaban, and the characteristics between the groups with recommendation for one or the other option differed significantly. ITEs resulted in recommendations that were significantly superior to usual (observed) treatment allocations in terms of absolute risk reduction, both separately for stroke and in the composite endpoint of stroke and major bleeding. When similar patients from routine data were selected (precision cohorts) for patients with a strong recommendation for one option or the other, those similar patients under the respective recommendation showed a significantly better prognosis compared with the alternative option. Many steps may still be needed on the way to clinical practice, but the principle of decision support developed from routine data may point the way toward future decision-making processes.