Daniel M. Keller, PhD

November 19, 2012

SAN DIEGO, California — Few urban adults adhere to targets in the Dietary Approaches to Stop Hypertension (DASH) diet, and nutrient intake differed between residents living in poverty and those not living in poverty, researchers reported at Kidney Week 2012: American Society of Nephrology 45th Annual Meeting.

Additionally, Deidra Crews, MD, assistant professor of medicine in the Division of Nephrology at Johns Hopkins University in Baltimore, Maryland, reported during a news conference that lower DASH adherence was associated with a higher risk for chronic kidney disease (CKD) among people in poverty.

Dr. Crews said that limited access to healthy foods and poor dietary habits are likely contributors to decreased estimated glomerular filtration rate (eGRF), higher albuminuria, and greater risk for end-stage renal disease (ESRD) among people in poverty or with low socioeconomic status. DASH diet adherence has been associated with better health; a lower risk for hypertension, type 2 diabetes, heart disease, and stroke; and eGFR decline. However, the relation of DASH diet adherence and disparities in CKD has been an open question.

Therefore, Dr. Crews and colleagues asked whether DASH diet adherence differed between people living in poverty or not and whether the relationship between DASH diet adherence and CKD differed between those groups. They performed a study using the prospective Healthy Aging in Neighborhoods of Diversity Across the Lifespan (HANDLS) population, a National Institute on Aging intramural study.

The HANDLS population under study (N = 2058 blacks and whites; n = 869 poverty, n = 1189 nonpoverty) was enrolled in 2004-2008 at ages 30 to 64 years. They came from 12 neighborhoods with socioeconomic and racial diversity in Baltimore, Maryland. The researchers defined poverty as a household income less than 125% of the federal poverty guideline of $23,562 for a family of 4.

DASH diet adherence was based on 3 macronutrients (saturated fat, total fat, and protein) and 6 micronutrients (cholesterol, fiber, magnesium, calcium, potassium, and sodium), with a total possible DASH score of 9.0 for meeting all the target values. Nutrient data were observational and were based on 24-hour dietary recall, ie, participants were not assigned to the diet. The primary definition of CKD was an eGFR of less than 60 mL/min/1.73m2.

Low Adherence to DASH Targets in Both Cohorts

Among the 2058 participants (mean age, 48 years; 57% black; 44% male; 42% poverty), both the poverty and nonpoverty cohorts had very low adherence to DASH targets, with median DASH scores of 1.64 and 1.84 out of 9.0, respectively (P < .001). The proportion adhering to the recommended diet (DASH score of 4.5 or greater) was similarly low — 4.5% vs 6.1% (P = .1). The poverty cohort had higher cholesterol intake and lower intake of fiber, magnesium, calcium, and potassium (all P < .001). The cohorts did not differ in their sodium intake, which was above the DASH target.

When stratified by diet adherence tertile, being male, black race, poverty, fewer years of education, lack of a healthcare provider, and tobacco use were all significantly associated with lower adherence.

The poverty group had a CKD prevalence of 5.6% vs 3.8% in the nonpoverty group (P = .05). DASH adherence was associated with CKD only among the poverty cohort (P interaction = .02).

Dr. Crews noted that strengths of the study were its examination of the role of diet in CKD disparities, well-balanced cohorts across a range of socioeconomic status, and its attention to adherence to the DASH guidelines, which is a well-established, effective dietary pattern. The study was limited by its cross-sectional analysis of the particular urban population, its dependence on dietary recall, and potential dietary confounders, such as preservatives.

The use of food frequency questionnaires to estimate nutrient intake is considered to be valid and is widely used in large epidemiologic and other studies in which it is not feasible to assign a diet or to measure actual intake, according to Kerri Cavanaugh, MD, MHS, assistant professor of medicine in the Division of Nephrology and medical director of the Vanderbilt Dialysis Clinic-Campus at the Vanderbilt University Medical Center in Nashville, Tennessee.

"It has known inherent limitations as it is self-report and asks patients to report their average intake of a variety of foods over a specified period of time, but is generally accepted to reflect general eating patterns and to inform hypothesis generating research to be later tested in other study designs," Dr. Cavanaugh wrote in an email interview with Medscape Medical News. "A limitation of a summary score (0-9) is that we are collapsing the elements of the DASH diet and thus unable to determine if specific parts, such as potassium, sodium, or fat, are more influential when considering the association between diet intake and risk of CKD. In summary, it is a valid approach but with limitations for interpretation."

She said the fact that both cohorts were far off the DASH targets reflects the situation that "we, as a population, are unlikely to generally consume diets that overall are similar to the composition of the DASH diet, regardless of poverty status." She suggested that there may have been some underreporting of total dietary intake, a known limitation with food frequency questionnaires. As evidence, although sodium intake in both cohorts was above the DASH target, it was still below what has been reported for the general US population.

In summary, Dr. Cavanaugh said the study suggests that among people in poverty, diet composition may be a contributor to the risk for CKD. "Perhaps if this were modified, among those with poverty, the risk may be reduced. Additional studies are needed to examine this hypothesis.

"The major clinical message is that if we agree that the DASH diet is recommended, everyone is doing a poor job at meeting the targets of that type of diet," she concluded.

Table: CKD Prevalence by DASH Adherence and Poverty Status*

DASH Adherence Tertile Poverty (Odds Ratio; 95% Confidence Interval) Nonpoverty (Odds Ratio; 95% Confidence Interval)
Lowest 7.1% (3.32; 1.26 - 8.73) 3.4% (0.77; 0.38 - 1.56)
Middle 6.3% (2.93; 1.06 - 8.13) 3.6% (0.81; 0.38 - 1.71)
Highest 2.3% (1.0 - reference) 4.4% (1.0 - reference)
P-trend .02 .5

*CKD: eGFR less than 60 mL/min/1.73 m2


There was no commercial funding for the study. Dr. Crews and Dr. Cavanaugh have disclosed no relevant financial relationships.

Kidney Week 2012: American Society of Nephrology 45th Annual Meeting. Abstract SA-OR002. Presented November 3, 2012.