The risk of all-cause mortality during the 5-year follow-up in patients with AF was lower for those with higher education level relative to those with primary education only. The positive effect increased with education level and persisted even after adjusting for multiple confounders.
Our findings are consistent with those of Patsiou et al. who found significantly lower all-cause mortality among highly-educated AF patients in a retrospective cohort study of 1,082 patients after a median 31-month follow-up. [19] Similarly, in the Cardiovascular Disease in Norway 2008–2012 project, Akerkar et al. prospectively followed 42,138 patients discharged after hospitalization with AF. They reported higher all-cause and cardiovascular mortality in groups with lower educational levels over a mean follow-up of 2.4 years. [20]
A Swedish cohort study examining 12,283 AF patients from 75 primary care centres, primarily in the Stockholm region, revealed a significantly lower relative mortality risk among patients with higher education levels after adjustment for several socioeconomic factors. Academic education was associated with a lower risk of heart failure in females and a lower risk of AMI in both sexes. [21]
Our sex-stratified analysis showed comparable mortality rates in males and females, despite higher mean age and greater stroke risk in females at index hospitalization. A systematic review reported similar results, although in studies with generally shorter follow-up periods. [22]
Our results support the impact of education level on all-cause mortality among individuals with AF independent of the primary hospitalization diagnosis. Lower education potentially contributes to a worse prognosis influenced by factors such as lower health literacy, limited disease awareness, decreased quality of life [23] [24], and less accessibility to advanced AF treatments such as catheter ablation. [7] [6] Poor adherence to prescribed medication, more common with lower education level, also plays a substantial role in adverse outcomes. [25]
Subpopulation analysis stratified by CCI score revealed a positive effect of education in all CCI groups. However, the effect on mortality risk was more pronounced in the low-risk CCI group compared to high-risk group.
While significant interaction HRs emerged when contrasting low/moderate with high comorbidity groups, those with higher education still showed reduced mortality risk. The positive effect of education was reduced in the early follow-up periods, but to a lesser degree in the long term.
Notably, neither CAD nor AMI significantly altered the association between education level and all-cause mortality. Heart failure exerted a significant impact on all-cause mortality risk in academic versus primary educated males. The stratification of CVE showed a significant interaction HR only in academically educated females.
When comparing patients stratified for cancer the analysis showed a time-dependent association. In the initial 2.5 years, the time-dependent mortality risk in cancer patients was significantly higher and the mortality did not differ with education level. In patients surviving beyond 2.5 years, a positive effect of the education was evident, although to a lesser extent than observed in the non-cancer population. A time-series study from the US, examining the data of 8.2 million individuals who died from cancer, showed a pronounced education gap in mortality regardless of the type of malignancy. This effect persisted even after policy changes that made screening programs and treatment more accessible to broader populations. [26] A study from Colombia reported consistent educational disparities in mortality trends even after the implementation of a more general health insurance system. [27] These findings suggest that worldwide data might be comparable, despite differences in healthcare systems.
Our results are in accordance with a recent Chinese prospective observation study, where stroke patients were followed for mortality and cardiovascular events for two years. This study showed a similar mortality reduction with higher education levels. The relative risk differences were higher in the Chinese population. The same paper concurrently published a meta-analysis which showed that comparable studies from western populations demonstrated a similar protective effect of education, albeit with shorter follow-up periods and significantly smaller populations. [28]
All observational studies inherently carry a potential for bias that must be considered when interpreting results. Despite utilizing data extracted from high-quality national registries, there is a risk of misclassification of diagnoses and coding errors. The impact of unknown confounders such as type of atrial fibrillation, the anticoagulants used, and variation in medical treatments cannot be disregarded.
Additionally, the absence of data from primary care settings and outpatient clinics introduces a challenge in determining whether patients who appeared to have incident AF might have had an AF diagnosis prior to hospital admission.
However, the strength of this study is the large number of patients and the extensive five-year follow up, both of which are based on data from high-quality national registries.