Mediterranean Diet Provides Cognitive Benefits for Hispanic, Latinx Adults

Researchers sought to assess the association of a Mediterranean diet with cognitive performance among community-dwelling Hispanic or Latinx adults.

Among community-dwelling, middle-aged and older Hispanic or Latinx adults, high adherence to a culturally tailored Mediterranean diet is associated with better cognitive performance, as well as with decreased 7-year learning and memory decline, according to study findings published in JAMA Network Open.

In the cohort study, researchers examined data from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) and the Study of Latinos–Investigation of Neurocognitive Aging (SOL-INCA) — with the latter being an HCHS/SOL ancillary study. Between March 2008 and June 2011 (visit 1), cognitive tests were administered in HCHS/SOL; in SOL-INCA, cognitive tests were given between October 2015 and

March 2018 (visit 2). All participants in the study had completed a dietary evaluation at visit 1, along with neurocognitive assessments at visits 1 and 2. All of the data were analyzed between September 2021 and May 2022.

Hispanic or Latinx ethnicity, which was self-reported by participants in HCHS/SOL, included Cuban, Central and South American, Mexican, Puerto Rican, and Dominican individuals. In both HCHS/SOL and SOL-INCA, neurocognitive tests were administered during face-to-face encounters by trained bilingual psychometrists in the participants’ preferred language.

Adherence to a Mediterranean diet was determined with use of the Mediterranean diet score (MDS), which was classified as low (MDS of 0-4 points), moderate (MDS of 5-6 points), or high (MDS of 7-9 points). The researchers used the mean of 2 24-hour dietary recalls to calculate the MDS.

At visit 1, all of the participants completed a 24-hour dietary recall of predefined food and nutritional categories. A second dietary recall was carried out about 30 days later. The consumption of various MDS components (ie, fruits, vegetables, legumes, cereals, fish, meat, and dairy products) was recorded and adjusted for total energy intake.

All participants received:

  • 1 point for each beneficial component (ie, fruits, vegetables, legumes, cereals, and fish) that was consumed at or above the sex-specific median
  • 1 point for each detrimental component (ie, meat and dairy products) consumed below the sex-specific median
  • 1 point for monounsaturated to saturated fat ratio consumed above the sex-specific median
  • 1 point for moderate alcohol consumption (2 drinks or less per day for men and 1 drink or none per day for women)

The neurocognitive tests administered included the Brief Spanish-English Verbal Learning Test (B-SELVT) Sum, B-SVELT Recall, word fluency, and the Digit Symbol Substitution Test (DSST). The cognitive change between visits 1 and 2 was computed by subtracting the cognitive score at visit 2 from the cognitive score at visit 1, then adjusting by the time that elapsed between the visits and the cognitive score at visit 1.

A total of 6321 individuals were included in the study. The mean age of the participants was 56.1±0.18 years at visit 1. Overall, 4077 of the participants were women. Results of the study showed that Mediterranean diet adherence-weighted frequencies were 35.8% among those in the low-adherence group, 45.4% among those in the moderate- adherence group, and 18.8% among those in the high-adherence group.

Per the fully adjusted model, z score–transformed cognitive scores at visit 1 in the high-adherence vs low-adherence groups were higher for B-SVELT Sum (95% CI, 0.02-0.20), B-SVELT Recall (95% CI, 0.07-0.25), and global cognition (95% CI, 0.04-0.16) tests.

At the mean 7-year follow-up, cognitive change in the high-adherence vs the low-adherence group was less pronounced for B-SVELT Sum (95% CI, 0.05-0.20) and

B-SVELT Recall (95% CI, 0.05-0.23), but not for word fluency, DSST score, or global cognition score.

Several limitations of the present study warrant mention. To begin, certain covariates were self-reported, thus introducing the potential for recall bias. Further, since the original MDS was used, the calculation of the score was based on the consumption of beneficial and detrimental foods above or below the sex-specific medians. Additionally, only 4 different tests were used to evaluate participants’ cognitive performance.

According to the researchers, “Future studies should examine whether the results prove similar across different racial and ethnic heritage.”

Disclosure: Some of the study authors have declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures. 


Moustafa B, Trifan G, Isasi CR, et al. Association of Mediterranean diet with cognitive decline among diverse Hispanic or Latino adults from the Hispanic Community Health Study/Study of Latinos. JAMA Netw Open. Published online July 14, 2022. doi:10.1001/jamanetworkopen.2022.21982