Sodium Intake and Cause-Specific Mortality Among Predominantly Low-Income Black and White US Residents

Abstract

Importance: Epidemiologic evidence regarding the outcomes of dietary sodium intake on mortality remains limited for low-income individuals, particularly Black people.

Objective: To investigate the associations of excessive dietary sodium with all-cause and cause-specific mortality among predominantly low-income Black and White Americans.

Design, setting, and participants: This cohort study included participants aged 40 to 79 years from the Southern Community Cohort Study who were recruited at Community Health Centers in 12 southeastern states from 2002 to 2009. Analyses were conducted between March 2022 and June 2023.

Exposures: Dietary sodium intake was assessed using a validated food frequency questionnaire at baseline.

Main outcomes and measures: Multivariable-adjusted Cox regression was used to estimate hazard ratios (HRs) and 95% CIs for mortality outcomes (all-cause, cardiovascular disease [CVD], coronary heart disease [CHD], stroke, heart failure, cancer, and other) associated with sodium intake. Nonlinear associations and population-attributable risk (PAR) of the mortality burden associated with excess sodium were further assessed.

Results: Among the 64 329 participants, 46 185 (71.8%) were Black, 18 144 (28.2%) were White, and 39 155 (60.9%) were female. The mean (SD) age at study enrollment was 51.3 (8.6) years for Black participants and 53.3 (9.3) years for White counterparts. Mean (SD) dietary sodium intake was 4512 (2632) mg/d in Black individuals and 4041 (2227) mg/d in White individuals; 37 482 Black individuals (81.2%) and 14 431 White individuals (79.5%) exceeded the current dietary recommendations of 2300 mg/d. During a median (IQR) follow-up of 13.8 (11.3-15.8) years, 17 811 deaths were documented, including 5701 from CVD. After adjustment for potential confounders, in Black individuals, HRs per 1000-mg increase in daily sodium intake were 1.07 (95% CI, 1.03-1.10) and 1.08 (95% CI, 1.02-1.14) for deaths from total CVD and CHD, respectively; while in White individuals, the corresponding HRs were 1.08 (95% CI, 1.02-1.14) and 1.13 (95% CI, 1.03-1.23). No significant associations were found for cancer mortality. PAR estimates suggest that sodium intake above the recommended threshold may account for 10% of total CVD, 13% of CHD, and 30% of heart failure deaths in this low-income southern population.

Conclusions and relevance: In this cohort study of 64 329 low-income Americans, nearly 80% of study participants consumed sodium exceeding the current recommended daily amount, which was associated with 10% to 30% of CVD mortality. Public health programs targeted to reduce sodium intake among this underserved population may be beneficial.

Figure 1.. Total Cardiovascular Disease Mortality in Relation to Dietary Sodium Intake: Subgroup Analysis by Baseline Characteristics

Hazard ratios (HRs) and 95% CIs for more than 6900 mg/d vs less than 2300 mg/d were estimated after adjustment for age, sex, education, income, marital status, medical insurance, smoking, physical activity, alcohol consumption, body mass index, total energy intake, and healthy eating index. Living alone included individuals who never married and those who were separated, divorced, widowed, or single. Physical activity was categorized by tertiles of the total metabolic equivalent of task hours per week. Alcohol consumption was defined as nondrinkers, 0 g/d; moderate drinkers, more than 0 but 28 or less g/d for men or more than 0 but 14 or less g/d for women; and heavy drinkers, more than 28 g/d for men or more than 14 g/d for women. Interaction was tested by the likelihood ratio test, comparing models with and without the multiplicative interaction term of sodium intake (continuous) × each stratum variable. Error bars represent the 95% CIs.

Figure 2.. Dose-Response Association of Dietary Sodium Intake With Total Cardiovascular Disease Mortality

Solid lines represent the hazard ratios and shaded areas represent the 95% CIs. The reference was set at 2300 mg/d, with 3 knots fitted at the 5th, 50th, and 95th percentiles. Participants with the highest 1% of sodium intake were excluded to reduce the noise of extreme outliers. All models were adjusted for age, sex, enrollment source, education, income, marital status, medical insurance, smoking, physical activity, alcohol consumption, body mass index, total energy intake, and healthy eating index.

Figure 3.. Population-Attributable Risk (PAR) for Cause-Specific Mortality by Dietary Sodium Intake Among Low-Income Black and White Americans

The calculated PARs represent a value of 0 if they were negative for mortality outcomes. Error bars represent the 95% CIs for the population-attributable risk estimates. CHD indicates coronary heart disease; CVD, cardiovascular disease; PAR, population-attributable risk.