Regarding: Poor long-term cardiovascular risk factor management after acute coronary syndrome: An observational cohort study

Abstract

Dear Editor,

We read with interest the study by Schubert et al. reporting suboptimal long-term cardiovascular risk factor monitoring and control after acute coronary syndrome (ACS). The authors should be commended for leveraging comprehensive electronic health records to characterize real-world care beyond the first year [1]. The findings of declining measurement rates and persistently low attainment of LDL-C and blood pressure targets are clinically relevant. However, several aspects warrant further consideration. In particular, ignoring myocardial infarction (MI) subtype (STEMI vs. NSTEMI), age, and the underlying sex distribution may introduce substantial confounding when evaluating sex differences in post-ACS management, potentially generating spurious associations or misleadingly attenuating true disparities.

Confounding by MI phenotype is critical in this setting. STEMI and NSTEMI represent distinct populations with markedly different epidemiological profiles. STEMI predominantly affects men (≈70%), whereas NSTEMI shows a more balanced sex distribution, with women being disproportionately represented among NSTEMI presentations. Importantly, NSTEMI patients are consistently less likely to receive intensive secondary prevention and achieve optimal adherence compared with STEMI [2]. As a result, analyses that pool both entities may artificially exaggerate apparent treatment gaps in women, simply reflecting their overrepresentation in a population that is globally undertreated [3]. Conversely, if sex disparities are more pronounced within NSTEMI itself, aggregation may dilute the true magnitude of within-group differences, masking clinically relevant inequities.

Age also represents a potential effect modifier. Women present with MI at substantially older ages than men across both STEMI and NSTEMI. This has several implications. Older patients are less likely to receive intensive therapies irrespective of sex, introducing treatment-intensity bias [4]. In addition, age is closely associated with a higher burden of comorbidities—such as chronic kidney disease or heart failure—that influence both therapeutic decisions and outcomes. Finally, the effect of sex on prognosis varies across age strata: younger women exhibit worse outcomes compared with men, whereas this difference attenuates or reverses in the elderly [5]. Failure to account for these interactions may therefore obscure meaningful sex-specific effects.

These combined mechanisms create a high risk of Simpson's paradox, whereby pooled analyses yield associations that differ from—or even contradict—those observed within clinically relevant subgroups [6]. For example, even if women receive less optimal care within both STEMI and NSTEMI, aggregation may suggest smaller or absent differences due to the unequal distribution of MI subtypes. Overall, failure to account for MI type and age is likely to underestimate true sex disparities. The concentration of women in NSTEMI and at older ages may distort the magnitude of differences that are in part attributable to inequities in care [7]. Robust analyses should therefore systematically account for MI phenotype and age, ideally through stratification or interaction modeling, to ensure valid interpretation of sex-based differences in secondary prevention.

In summary, this study highlights important gaps in secondary prevention after ACS. However, the findings point to a multifactorial issue possibly involving healthcare organizations, therapeutic inertia, and case-mix complexity. Careful consideration of infarct type, age, and patient characteristics is essential to avoid biased interpretation of sex differences and to guide effective interventions.

Laurent Fauchier: Conceptualization; writing—original draft. Mickael Guglieri: Writing—review and editing; conceptualization; validation. Thibaud Genet: Writing—review and editing; conceptualization; validation.

L.F.: consultant or speaker fees (small amounts) for AstraZeneca, Bayer, BMS/Pfizer, Boehringer Ingelheim, Boston Scientific, Medtronic, Novo Nordisk, and Zoll outside the scope of this work. Other authors declare no conflicts of interest.

None.