Introduction: Greater adherence to a DASH diet pattern is linked to lower cardiovascular disease (CVD) risk. However, it is unclear whether a brief DASH screener, validated in a healthy, nationally representative population, demonstrates predictive validity in secondary prevention settings. Hypothesis: We examined the associations between 6-month (mo) Δ in DASH screener score and CVD biomarkers at 1-year (yr) and 2-yr follow-up in the SAMMPRIS trial. We hypothesized that improved 6-mo Δ in DASH screener scores were associated with higher high-density lipoprotein cholesterol (HDL-C), and lower systolic blood pressure (SBP), diastolic blood pressure, and low-density lipoprotein cholesterol, at 1-yr and 2-yr follow-ups, and that the associations were not modified by treatment. Methods: This secondary analysis of the SAMMPRIS trial included a subset of 244 participants who did not develop stroke or die ≤ 30 days after enrollment and responded to 8 brief nutrition questions used to calculate a DASH adherence score at both baseline and 6-mo. The screener captures 8 DASH components. Each component was assigned a corresponding weight and summed to obtain a total score (0-100). Higher values indicated better adherence. Multivariable linear regressions assessed associations between 6-mo Δ in DASH screener scores and CVD biomarkers at 1- and 2-yr follow ups, adjusted for covariates. Effect modification by treatment arm was tested, and inverse-probability-of-censoring weight was used to account for informative dropout/death. Results: Participants were on average 61.0(10.9) years, 75% White, 39% female, with a BMI of 30.3(6.1) kg/m 2 and a baseline screener score of 52.5(12.2) (see Table 1 ). Preliminary results show that 6-mo Δ in the screener score comparably increased in both the aggressive medical management plus stenting arm (+3.2) and the aggressive medical management only arm (+4.8). At 1-yr follow up, associations differed by treatment, and 6-month Δ in DASH screener score was associated with lower SBP (ß =-0.23, p=0.04) and higher HDL-C (ß =0.15, p=0.03) in the aggressive medical management plus stenting arm only. At 2-yr follow up, no associations were detected between 6-mo Δ in DASH screener score and CVD biomarkers in either arm (see Table 2 ). Conclusions: The DASH screener demonstrated short-term predictive validity with some CVD biomarkers in a high-risk clinical population who underwent aggressive medical management of vascular risk factors plus stenting.
Yao et al. (Tue,) studied this question.