Clearing up misunderstandings: addressing issues in recent cardiovascular risk review

Le résumé fourni par la source

This commentary refers to ‘Cardiovascular disease risk communication and prevention: a meta-analysis’, by M. Bakhit et al., https://doi.org/10.1093/eurheartj/ehae002 and the discussion piece ‘Cardiovascular disease risk communication alone is not enough: why a recent review may be misleading’, by S. Griffiths et al., https://doi.org/10.1093/eurheartj/ehae529. Thank you for your detailed and thoughtful feedback on our recent publication regarding cardiovascular disease (CVD) risk communication. We appreciate the authors’ acknowledgment that ‘the review conduct was generally appropriate’, but we note the difference in interpretation and appreciate the opportunity to address the concerns raised and provide clarification based on our study’s findings. We agree that ‘risk communication alone has limited impact on behaviour’. However, our review looked at whether risk communication added to advice about CVD risk management changed behaviour, that is, comparisons of (i) CVD risk management only vs. (ii) CVD risk management plus risk communication. Research has shown that communicating personalized CVD risk can influence patient behaviour and trigger sustainable changes1–3 for improved health outcomes.4,5 We used the term ‘may’ in our statement ‘Knowledge of quantifiable CVD risk may improve health outcomes and trigger behavioural change’ (p.998) to indicate potential rather than certainty. Given the variability in individual responses to risk information, it is appropriate because not every patient will react the same way to knowledge of their CVD risk; some studies show positive outcomes, while others show minimal impact, reflecting the complex interplay of factors influencing behaviour. Our study examined both intentions and actual outcomes regarding medication or diet changes. We found that CVD risk communication effectiveness varies between primary and secondary prevention groups and should be tailored to individual risk factors and health conditions. Unlike the referenced review provided by the authors that found no significant dietary improvement and did not perform meta-analyses, we meta-analysed some studies and reported on non–meta-analysable studies separately, explaining the differing findings. On page 1002, we clearly reported our intended analyses and the reasons for deviations from the published trial protocol. Since subgroup analyses were not feasible, we reported the relevant data in our ‘Supplementary data online, Table S2’. We want to clarify that in studies where a subgroup of patients had CVD, we classified them accordingly and extracted data from this subgroup only. We disagree with the authors regarding our summary assessments of the risk of bias, which are reported in the Supplementary data online, Figure S1. If referring to the ‘Overall’ column, we highlight that we used the Cochrane ROB tool 1, not 2. We have clearly highlighted in the Results section (page 1003) that our studies ‘generally had low risk of selection bias with appropriate random sampling, low attrition rates, and complete reporting of intended outcomes’. We also noted a potential for publication bias in the same section. We appreciate the opportunity to clarify these points and believe that our study provides valid results and helpful insights into the benefits of CVD risk communication strategies to inform practice. P.G. was contracted by the National Heart Foundation to review various issues on cardiovascular risk, including some of the published systematic review as a component.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.