Risk of cardiovascular disease among cancer survivors: Protocol of a pooled analysis of population-based cohort studies

INTRODUCTION: Cancer and cardiovascular disease remain leading causes of death and disability worldwide, which places a heavy burden on public health systems and causes widespread suffering. Because these entities have highly overlapping risk factors, including hyperlipidemia, hypertension, diabetes, obesity, smoking and other lifestyle factors, many studies have reported that they have similar etiological mechanisms. Accumulating evidence indicates that there is an increased risk of cardiovascular disease among cancer survivors compared with the general population. However, whether cancer is associated with an increased risk of cardiovascular disease remains controversial. METHODS AND ANALYSIS: We will conduct and report the meta-analysis strictly based on the Cochrane Handbook for Systematic Reviews and the Meta-analysis of Observational Studies in Epidemiology guidelines combined with the Preferred Reporting Items for Systematic Reviews and Meta-analysis for Protocols (PRISM-P). This meta-analysis was registered with PROSPERO (registration number CRD42022307056). We will search for studies published from database inception to December 1, 2021, regardless of language or date, in three electronic databases (PubMed, EMBASE, and Cochrane Library) to identify and appraise cohort studies examining the relationship between cancer and subsequent cardiovascular disease risk. The literature screening, inclusion and data extraction will be conducted independently by two investigators using pre-designed standardized data extraction forms. A senior investigator will be consulted in cases of disagreement. We will assess risk of bias in the included cohort studies using the Newcastle–Ottawa Scale (NOS). Quantitative synthesis will be conducted using a random-effects model. To explore potential sources of heterogeneity, we will carry out multiple sensitivity analysis, meta-regression and subgroup analysis according to baseline characteristics. Publication bias will be evaluated through visual inspection of funnel plot asymmetry as well as by Begg's rank correlation test and Egger's weighted linear regression test.