Dr Shu Ren

Objectives: To examine the effectiveness of different strategies for recruiting participants for a large Australian randomised controlled trial (RCT), the Australian Study for the Prevention through Immunisation of Cardiovascular Events (AUSPICE). Design, setting, participants: Men and women aged 55¿60 years with at least two cardiovascular risk factors (hypertension, hypercholesterolaemia, overweight/obesity) were recruited for a multicentre placebo-controlled RCT assessing the effectiveness of 23-valent pneumococcal polysaccharide vaccine (23vPPV) for preventing cardiovascular events. Methods: Invitations were mailed by the Australian Department of Human Services to people in the Medicare database aged 55¿60 years; reminders were sent 2 weeks later. Invitees could respond in hard copy or electronically. Direct recruitment was supplemented by asking invitees to extend the invitation to friends and family (snowball sampling) and by Facebook advertising. Main outcome: Proportions of invitees completing screening questionnaire and recruited for participation in the RCT. Results: 21¿526 of 154¿992 invited people (14%) responded by completing the screening questionnaire, of whom 4725 people were eligible and recruited for the study. Despite the minimal study burden (one questionnaire, one clinic visit), the overall participation rate was 3%, or an estimated 10% of eligible persons. Only 16% of eventual participants had responded within 2 weeks of the initial invitation letter (early responders); early and late responders did not differ in their demographic or medical characteristics. Socio-economic disadvantage did not markedly influence response rates. Facebook advertising and snowball sampling did not increase recruitment. Conclusions: Trial participation rates are low, and multiple concurrent methods are needed to maximise recruitment. Social media strategies may not be successful in older age groups. Trial registration: Australian New Zealand Clinical Trials Registry, ACTRN12615000536561.

Background: The pneumococcal polysaccharide vaccine (PPV) has been associated with reduced risk of cardiovascular events in human observational studies. Animal studies suggest that the phosphorylcholine epitope in the Streptococcus pneumoniae cell wall is structurally similar to oxidized low-density lipoprotein (oxLDL), hence PPV induces the production of antibodies that cross-react with anti-oxLDL and may cause regression of atherosclerotic plaque. We set out to determine the strength of association between PPV administration and reduction in cardiovascular events. Methods: A longitudinal, population-based cohort study of older Australians, from the Hunter Community Study, with up to 11 years of follow-up. We included participants aged = 65 years at baseline (2004¿2008), without a history of cardiovascular disease (CVD). History of PPV administration at baseline was the main exposure of interest. "Total number of hospital bed-days with CVD primary diagnosis" was one of the main outcomes measured. Models were adjusted for age, diabetes, alcohol intake, and smoking status. Influenza vaccine was the control exposure used and fracture bed-days was the control outcome used, to investigate the potential for residual confounding. Results: 91 of the total 1074 participants (mean age = 72, male = 45%) experienced a CVD event during follow-up. PPV (regardless of influenza vaccine) was associated with a significant reduction in CVD bed-day, (n = 863, incident rate ratio, IRR = 0.65, 95%CI: 0.45¿0.94, p = 0.02), but influenza vaccine (regardless of PPV) was not (n = 864, IRR = 0.86, 95%CI: 0.54¿1.35, p = 0.51). Furthermore, PPV adjusted for influenza vaccine remained associated with CVD bed-days (IRR = 0.64, 95%CI: 0.43¿0.96, p = 0.03) but was not associated with fracture bed-days (IRR = 0.75, 95%CI: 0.28¿2.00, p = 0.56). Conclusion: PPV demonstrated a 35% reduction in CVD bed-days. This finding was robust to residual confounding, using a control exposure and a control outcome, eliminating the concern for healthy-user bias. A large double-blinded placebo-controlled RCT is underway to confirm our finding and to explore the proposed mechanism of action (ACTRN12615000536561).