Dietary modification is a foundational component of hypertension management. Excessive intake of red and processed meats has been linked in prior literature to higher risk of hypertension and cardiovascular outcomes. This study investigates how red meat consumption is distributed across socioeconomic groups among adults with hypertension in Iran, using a wealth-related inequality framework. The authors applied Wagstaff decomposition to quantify the degree of inequality and to identify which factors contribute most to any pro-rich or pro-poor concentration of excessive consumption.
The analysis used data from the 2021 Iranian STEPwise Approach to non-communicable disease (NCD) Risk Factor Surveillance (STEPS) survey. The STEPS survey is a nationwide, cross-sectional surveillance effort that sampled adults aged 18 years and older across all 31 provinces via systematic cluster random sampling. Fieldwork began in early 2020, paused for the COVID-19 pandemic, and resumed in 2021. The original household sample comprised 27,874 adults; a subpopulation of individuals meeting the study definition of hypertension was selected for this analysis.
From 9,572 hypertensive participants, 548 were excluded for missing key variables (including wealth index, education, comorbidity, or meat consumption), yielding a final analytic sample of 9,024 hypertensive adults. The analysis used survey sample weights that adjusted for non-response and provided post-stratification by province, residential area, sex, and age to produce nationally representative estimates.
The primary outcome was excessive red meat consumption. Participants were asked how often they consumed red meat (examples provided: beef, veal, lamb, goat, mutton, camel). Response categories ranged from never/less than once a month to three or more times per day. Those reporting consumption of red meat four or more times per week were classified as excessive consumers. This threshold was chosen to capture frequent intake consistent with evidence linking high red meat consumption to adverse health outcomes.
Covariates included age (18–39, 40–59, ≥60 years), sex (male/female), marital status (married vs single which included never-married, divorced, widowed), area of residence (urban/rural), education (no formal education, 1–6 years, 7–11 years, ≥12 years), job status (employed, unemployed, unpaid work, retired), and comorbidity (presence of at least one of diabetes, hypercholesterolemia, cardiovascular disease, or cancer). The household wealth index was derived using principal component analysis (PCA) of assets and housing characteristics and categorized into quintiles.
Inequality in excessive red meat consumption was assessed visually with the concentration curve (CC) and summarized numerically using the concentration index (CI). A positive CI indicates that the outcome is more common among wealthier individuals (pro-rich concentration); a negative CI indicates concentration among the poorer (pro-poor). Given the binary outcome, the Wagstaff normalization was applied to ensure CI values were interpretable within the –1 to +1 range.
To decompose the observed wealth-related inequality, the authors used Wagstaff decomposition. A logistic regression model for the binary outcome was fitted and average marginal effects were extracted for use in the decomposition. The decomposition expresses the overall CI as a weighted sum of the concentration indices of explanatory variables; weights are proportional to the elasticities (marginal effects) of the outcome with respect to each covariate. The adjusted contribution method by Yiengprugsawan et al. was applied so the explained components align with the direction of the overall CI. Multicollinearity checks reported a maximum variance inflation factor (VIF) of 2.23, indicating no problematic collinearity. All analyses were performed in Stata 17.
The analytic sample of 9,024 hypertensive adults had a mean age of 56.1 years (95% CI: 55.8, 56.5). Overall, 16.5% of participants met the study definition of excessive red meat consumption (four or more times per week).
A statistically significant pro-rich wealth-related inequality in excessive red meat consumption was observed (CI = 0.179, p < 0.001), indicating that excessive consumption was concentrated among wealthier individuals. The Wagstaff decomposition attributed the largest single contribution of the explained component to higher educational attainment (≥12 years), which accounted for 20.6% of the explained inequality. Other notable contributors included unpaid work (6.8%), unemployment (4.4%), and being aged 40–59 years (4.2%). Collectively, the included determinants explained 29.4% of the observed inequality; the remaining portion of the CI was unexplained by the variables in the model.
Descriptive comparisons between non-excessive and excessive consumers showed differences by age distribution: the excessive consumption group had a higher proportion aged 40–59 years (47.2% vs 42.8%) and a lower proportion aged ≥60 years (40.4% vs 43.9%). Sex distributions and area of residence did not differ significantly between consumption groups in the reported comparisons.
This analysis demonstrates a measurable pro-rich concentration of frequent red meat intake among adults with hypertension in Iran. The finding that higher educational attainment was the largest contributor to the explained inequality may reflect complex relationships between education, dietary preferences, affordability, and access to diverse food options in this setting. The result that only 29.4% of the inequality was explained by measured factors underscores that additional socioeconomic, cultural, market, or behavioral determinants not captured in the survey may be important.
From a public health perspective, the authors suggest that interventions addressing dietary risk among hypertensive adults should consider socioeconomic gradients. Targeted nutrition education, affordability and access strategies, and context-sensitive policies may be warranted to reduce cardiovascular risk and improve equity in dietary behaviors.
The analysis used complete-case methods and excluded participants with missing data on key variables; this approach may introduce selection bias if excluded individuals differ systematically. Cross-sectional STEPS data do not allow causal inference. The decomposition explained less than one third of the observed inequality, indicating residual unexplained contributors. The study reported ethics approval from the Endocrinology & Metabolism Research Institute at Tehran University of Medical Sciences and noted that data access for secondary research requires formal application to the NIHR as per the source.
Among Iranian adults with hypertension, excessive red meat consumption (≥4 times/week) was present in 16.5% and was significantly concentrated among wealthier individuals (CI = 0.179). Educational attainment was the largest measured contributor to this pro-rich inequality. The findings support the need for socioeconomic-aware nutrition policies and targeted interventions to address dietary risk and promote equity in cardiovascular risk reduction.