-
On the role of marginal confounder prevalence – implications for the high-dimensional propensity score algorithm
We apply the law of total probability in conjunction with an established bias formula to derive and illustrate relative bias boundaries with respect to marginal confounder prevalence.
Retrospective analysis of trends in dispensing of long-acting non-tamper-deterrent oxycodone near the USA-Canada border
Our population-based analysis of all long-acting oxycodone dispensed in Canadian pharmacies near the Canada–US border found no evidence of behaviour suggestive of large-scale trafficking of generic non–tamper-resistant formulations after their introduction in Canada. However, Canadian clinicians and pharmacists should remain cautious in their prescribing and dispensing because of the high potential for misuse and abuse of this medication.
Q13-03AConfounding, effect modification and the odds ratio: Common misinterpretations
We review issues with regard to the use of the odds ratio and collapsibility, and demonstrate the relation between collapsibility and effect modification.
Higher potency statins and the risk of new diabetes: multicentre, observational study of administrative databases
Patients using high potency statins after a heart attack or stroke have a slightly greater risk of developing diabetes, compared to patients using low potency statins.
Q12-02Atypical antipsychotics and hyperglycemic emergencies: multicentre, retrospective cohort study of administrative data
Risk for hyperglycemic emergencies is low after initiation of antipsychotics, but patients with preexisting diabetes may be at greater risk. The risk appeared lower with the use of quetiapine in older patients, but the clinical significance of the findings requires further study.