Pharmacological prevention of ischemic stroke and TIA : epidemiological aspects and how to improve treatment

Abstract

Background: Medications can prevent stroke but are not used optimally. The overarching aim of this thesis was to study medication use in patients with previous ischemic stroke or transient ischemic attack (TIA) and in all patients with atrial fibrillation. Socioeconomic and demographic factors such as sex, education, and income have been associated with differences in medication use after stroke. Understanding these associations better may help in understanding reasons for suboptimal medication use. In the chronic setting, patients with a previous stroke are followed in primary care in Sweden. Primary care is thus an important target for improving medication use. All patient visits in primary care require that a diagnosis is recorded by the doctor in the patient’s electronic medical record. This “recording” of diagnoses has been hypothesized as a potential quality indicator, but the utility has not yet been proven. Also, the association between diagnosis recording and medication use has not been studied. Audit & feedback is a commonly used approach to achieve changes in behavior in healthcare personnel. Changing the prescribing and motivating behavior of primary care doctors vis-à-vis stroke/TIA and atrial fibrillation patients could potentially increase medication use.

Methods: All the studies in this thesis were registry based and have included patients ≥18 years of age from Region Stockholm. The outcome of all studies has been medication use. By using the Swedish National Prescribed Drug Register (NPDR), we were able to study medication dispensation to patients as a marker of medication use. Study I used cross-linked data from the VAL database (see below), NPDR, and Statistics Sweden. Studies II-IV used data from the local healthcare administrative database for Region Stockholm, the VAL database. Data in VAL is identical to that found in the National Patient Register (NPR) and since 2010 also the NPDR. In study I we explored the association between medication use and socioeconomic and demographic factors 9-12 months after ischemic stroke/TIA. Study II explored the association between diagnosis recording in primary care and medication use for the diagnoses stroke/TIA and acute coronary syndrome. Studies III and IV tested if an audit & feedback intervention in primary care could improve medication use and diagnosis recording in patients with ischemic stroke/TIA (III) or atrial fibrillation (IV).

Results/conclusions: Use of recommended preventive medications in Region Stockholm has increased over time in both patients with prior ischemic stroke/TIA and patients with atrial fibrillation. Although statin use has increased, statins are still the secondary preventive medication class which is used the least after ischemic stroke/TIA. The sex gap in statin use after ischemic stroke/TIA has persisted over time and future interventions should target improving statin use particularly in women. High income was associated with being dispensed more statins, anticoagulants, and antiplatelets 9-12 months after ischemic stroke/TIA. Having a diagnosis recorded in primary care was associated with greater use of antithrombotics and statins in ischemic stroke/TIA, and acute coronary syndrome. Also, recorded atrial fibrillation patients used more anticoagulants. An audit and feedback intervention did not improve the utilization of preventive stroke medications in primary care.