To investigate if beliefs about medicines affect the time to the initiation of cardiovascular preventive medications during a 3 year follow-up period.

A questionnaire and register-based cohort study.

Primary care in Sweden, in which people 40, 50 and 60 years old underwent risk factor screening and individual health promotion within the Västerbotten intervention programme (VIP).

People at low/medium risk of cardiovascular disease (CVD) according to the risk factor screening were included in the VIsualiZation of asymptomatic Atherosclerotic disease for optimum cardiovascular prevention—a population-based Pragmatic Randomised Open Blinded End-point trial (PROBE) nested in the Västerbotten Intervention Programme (VIPVIZA), aiming at improved primary prevention of CVD. People participating in the VIPVIZA 3 year follow-up (n=3167 (89.7%)), receiving the Beliefs about medicines questionnaire (BMQ) (n=2314 (73.1%)) and with complete answers to at least one subscale in the BMQ general (n=2258 (97.6%)) were included. Moreover, only those 60 years old at baseline (n=2073 (58.7%)) and without antihypertensive and/or lipid-lowering drugs (n=1769 (50.1%)) 6 months prior to inclusion in the VIPVIZA trial were included. Accordingly, the final study population comprised 888 people without antihypertensive medicines and 1185 without lipid-lowering drugs, respectively.

The primary outcome was time to the binary event of initiating antihypertensives or lipid-lowering agents, identified within the time frame from inclusion in the VIPVIZA study until the study participants’ 3 year follow-up visit. General beliefs about medicines were assessed according to the BMQ. Cox proportional hazards models, adjusted for sex, were conducted to investigate primary outcome.

Participants with stronger general beliefs about medicines being overused had significantly longer time to initiation of antihypertensive drugs in the control group (HR 0.91; 95% CI 0.84 to 0.996) but not in the intervention group (HR 1.05; 95% CI 0.95 to 1.16). No significant associations were found between beliefs about medicines and initiation of lipid-lowering treatment.

A more negative perception of drugs being overused was significantly associated with delayed initiation of antihypertensive drug treatment. Our results suggest that the VIPVIZA intervention may overbridge negative perceptions and affect the initiation of antihypertensive medications in a positive manner.

NCT01849575 (date of registration: 8 May 2013).

The use of direct oral anticoagulants (DOACs) as stroke prophylaxis in patients with non-valvular atrial fibrillation (NVAF) has increased steadily since the introduction in 2011. In Sweden today, more patients are treated with DOACs than with warfarin. However, it is not shown that an increased proportion of DOAC prescriptions correlates to lower event rates of stroke and systemic embolism.

This study aims to investigate whether the increased prescription of DOACs in Sweden correlates with lower event rates for all-cause stroke, systemic embolism and bleeding complications, using real-life data for the whole NVAF population.

Nationwide retrospective register study.

Data were obtained from the Swedish National Patient Registry, covering patients aged 18 years or older with NVAF, between 1 January 2014, and 31 December 2017. Exposure to oral anticoagulants was determined based on pharmaceutical data, calculating treatment duration by the number of pills dispensed and the prescribed daily usage rate. Baseline characteristics and endpoints were collected from hospital administrative registers using International Classification of Diseases, 10th edition (ICD-10) codes.

All patients with NVAF were identified using ICD-10 codes during the study period. Entry criteria included having a first recorded atrial fibrillation diagnosis after 1 January 2014 or being previously diagnosed with atrial fibrillation before 2014 but still receiving care after this date.

The outcomes were event rates (per 100 person-years) of ischaemic stroke, systemic embolism, all-cause stroke, major bleeding and intracranial bleeding (including haemorrhagic stroke). Event rates were calculated and compared across the study period using Cox proportional hazard models.

In the total NVAF population, the proportional decrease in event rates (per 100 treatment years) in 2017 compared with 2014 was ischaemic stroke 24% (1.70–1.30), all-cause mortality 4% (9.40–9.00), all-cause stroke 24% (2.10–1.60) and all-cause stroke and systemic embolism 23% (2.20–1.70). During the same time, the proportion of major bleeding and intracranial bleeding rates, including haemorrhagic stroke, also decreased: 5% (2.00–1.90), 6% (0.68–0.64) and 17% (0.30–0.25), respectively. DOACs use increased from 4.1% to 28.3% in the total population and from 22.7% to 60.9% in newly diagnosed patients.

In the initial years following the introduction when DOAC uptake in the population was high, an increasing proportion of Swedish NVAF patients receiving DOACs was accompanied by lower event rates of all-cause stroke and systemic embolism, ischaemic stroke and all-cause mortality, intracranial bleeding and major bleeding, highlighting the improved risk-benefit balance of DOACs in stroke prophylaxis.