Abstract
This cross-sectional observational study examined the association between dietary copper intake and cognitive function in American older adults, using data from the 2011 to 2014 National Health and Nutrition Examination Survey (NHANES). Analyzing a total of 2420 participants, dietary copper intake was determined by averaging two 24-h dietary recalls, whereas cognitive function was assessed by the Digit Symbol Substitution Test (DSST), the Animal Fluency Test (AFT), a Consortium to Establish a Registry for Alzheimer’s disease (CERAD) subtest and global cognition Z score. Multivariate linear regression models were used to explore the association between copper levels and cognitive function. Higher copper intake was associated with higher cognitive scores. In the fully adjusted model, compared to the lowest quartile (Q1), the highest quartile (Q4) of copper intake was associated with related to higher cognitive scores (DSST: β = 3.80, 95% CI 1.90,5.70; AFT: β = 1.23, 95% CI 0.48,1.99; CERAD-IRT: β = 0.58, 95% CI − 0.06,1.22; CERAD-DRT: β = 0.47, 95% CI 0.15,0.80; Z score: β = 0.20, 95% CI 0.10,0.29), particularly in participants with a history of stroke. Multivariate smooth spline analysis revealed that dietary copper intake was related to DSST, AFT and Z score in an inverted L-shaped nonlinear manner. The inflection point of copper was 1.63 mg/day for DSST, 1.42 mg/day for AFT and 1.22 mg/day for the Z score. Further longitudinal research is necessary to substantiate these findings.
Subject terms: Health care, Medical research, Neurology
Introduction
With the increase in the aging global aging population, the prevalence of cognitive impairment is also escalating1. All forms of dementia, ranging from mild cognitive impairment (MCI) to Alzheimer’s disease (AD), are marked by cognitive decline and are increasingly emerging as a significant global public health challenge2,3. According to the Global Burden of Disease Study, the number of people with dementia is expected to reach 152.8 million by 20504. Thus, researching protective factors for cognitive abilities is vital for preventing cognitive impairment5. The role of micronutrients in reducing cognitive decline has received considerable attention in recent years6. Particularly in the elderly population, vitamin and mineral supplementation is thought to potentially help maintain cognitive function and even prevent dementia. For example, the imbalance of certain micronutrients such as zinc, selenium, and copper in the brain is considered thought to be associated with cognitive decline and the development of neurodegenerative diseases7–9.
Copper, an essential trace element, is known to be a vital component for various physiological functions, including the proper development and functioning of the nervous system10. It is a key cofactor for several enzymes involved in neurotransmitter synthesis, cellular energy production, and antioxidant defenses. However, the relationship between copper and cognitive function is complex and not yet fully understood. Though, copper is necessary for proper functioning of the brain, and deficiencies can lead to neurological disorders, excessive copper can be toxic, leading to oxidative stress and neurodegeneration3,11. Therefore, this study aimed to investigate the relationship between dietary copper intake and cognitive function in individuals aged ≥ 60 years by analyzing publicly available data from the National Health and Nutrition Examination Survey (NHANES) for the years 2011 to 2014. Furthermore, we evaluated the dose–response relationship between dietary copper consumption and cognitive function.
Results
Participant characteristics at baseline
During the 2011–2014 cycle, 19,931 individuals participated in the NHANES. Individuals aged < 60 years (n = 16,299), those with missing cognitive data (n = 698), those with missing dietary data on copper intake (n = 221), and those with missing covariates (n = 293) were excluded for the study. Thus, a final sample of 2,420 participants was included in this study. Figure 1 illustrates the selection process in detail.
Fig. 1.
Flow chart of the screening and enrollment of study participants.
Table 1 shows the general characteristics of the participants according to their dietary copper levels. Quartile analysis of copper was conducted to categorize the participants into four groups: Q1 (< 0.76 mg/day), Q2 (0.76–1.04 mg/day), Q3 (1.05 -1.43 mg/day), and Q4 (≥ 1.44 mg/day). The mean age of the participants was 69.3 ± 6.7 years, with 1189 (49.1%) males and 1214 (50.2%) Non-Hispanic Whites. Moreover, compared to individuals with low copper intake, those with higher copper intake exhibited a greater likelihood of being Non-Hispanic White and male, as well as possessing higher family income, and lower rates of smoking. Additionally, these individuals demonstrated higher scores on the DSST, AFT, CERAD-IRT, CERAD-DRT and Z score, with respective averages of 50.5 ± 16.1, 18.1 ± 5.6, 19.6 ± 4.6, 6.3 ± 2.2,and 0.17 ± 0.75 in the highest copper intake.
Table 1.
Characteristics of the study population, National Health and Nutrition Examination Survey (NHANES) 2011–2014 (N = 2420).
| Variables | Copper intake, mg/d | |||||
|---|---|---|---|---|---|---|
| Total | Q1(< 0.76) | Q2(0.76–1.04) | Q3(1.05–1.43) | Q4(≥ 1.44) | p-value | |
| Number of participants | 2420 | 605 | 604 | 604 | 607 | |
| Age (years) | 69.3 ± 6.7 | 69.5 ± 6.5 | 69.8 ± 6.9 | 69.4 ± 6.8 | 68.7 ± 6.6 | 0.021 |
| Gender, n (%) | < 0.001 | |||||
| Male | 1189 (49.1) | 228 (37.7) | 272 (45.0) | 318 (52.6) | 371 (61.1) | |
| Female | 1231 (50.9) | 377 (62.3) | 332 (55.0) | 286 (47.4) | 236 (38.9) | |
| BMI (kg/m2) | 29.2 ± 6.4 | 29.9 ± 6.6 | 29.3 ± 6.5 | 29.4 ± 6.5 | 28.3 ± 6.2 | < 0.001 |
| Race/ethnicity, n (%) | < 0.001 | |||||
| Non-Hispanic white | 1214 (50.2) | 285 (47.1) | 307 (50.8) | 304 (50.3) | 318 (52.4) | |
| Non-Hispanic black | 567 (23.4) | 183 (30.2) | 153 (25.3) | 120 (19.9) | 111 (18.3) | |
| Mexican American | 206 (8.5) | 45 (7.4) | 47 (7.8) | 61 (10.1) | 53 (8.7) | |
| Others | 433 (17.9) | 92 (15.2) | 97 (16.1) | 119 (19.7) | 125 (20.6) | |
| Education level (year), n (%) | < 0.001 | |||||
| < 9 | 245 (10.1) | 88 (14.5) | 54 (8.9) | 56 (9.3) | 47 (7.7) | |
| 9–12 | 888 (36.7) | 265 (43.8) | 240 (39.7) | 208 (34.4) | 175 (28.8) | |
| > 12 | 1287 (53.2) | 252 (41.7) | 310 (51.3) | 340 (56.3) | 385 (63.4) | |
| Marital status, n (%) | < 0.001 | |||||
| Married or living with a partner | 1412 (58.3) | 313 (51.7) | 330 (54.6) | 375 (62.1) | 394 (64.9) | |
| Living alone | 1008 (41.7) | 292 (48.3) | 274 (45.4) | 229 (37.9) | 213 (35.1) | |
| Family income, n (%) | < 0.001 | |||||
| Low | 703 (29.0) | 237 (39.2) | 177 (29.3) | 152 (25.2) | 137 (22.6) | |
| Medium | 929 (38.4) | 221 (36.5) | 250 (41.4) | 238 (39.4) | 220 (36.2) | |
| High | 788 (32.6) | 147 (24.3) | 177 (29.3) | 214 (35.4) | 250 (41.2) | |
| Smoking status, n (%) | < 0.001 | |||||
| Never | 1176 (48.6) | 299 (49.4) | 308 (51) | 293 (48.5) | 276 (45.5) | |
| Former | 945 (39.0) | 202 (33.4) | 222 (36.8) | 248 (41.1) | 273 (45) | |
| Current | 299 (12.4) | 104 (17.2) | 74 (12.3) | 63 (10.4) | 58 (9.6) | |
| Alcohol status, n (%) | < 0.001 | |||||
| Never | 351 (14.5) | 99 (16.4) | 95 (15.7) | 97 (16.1) | 60 (9.9) | |
| Former | 390 (16.1) | 129 (21.3) | 107 (17.7) | 78 (12.9) | 76 (12.5) | |
| Current | 1679 (69.4) | 377 (62.3) | 402 (66.6) | 429 (71) | 471 (77.6) | |
| Hypertension, n (%) | 0.002 | |||||
| Yes | 1269 (52.4) | 349 (57.7) | 328 (54.3) | 303 (50.2) | 289 (47.6) | |
| No | 1151 (47.6) | 256 (42.3) | 276 (45.7) | 301 (49.8) | 318 (52.4) | |
| Diabetes, n (%) | 0.011 | |||||
| Yes | 565 (23.3) | 162 (26.8) | 152 (25.2) | 134 (22.2) | 117 (19.3) | |
| No | 1855 (76.7) | 443 (73.2) | 452 (74.8) | 470 (77.8) | 490 (80.7) | |
| Coronary heart disease, n (%) | 0.668 | |||||
| Yes | 222 (9.2) | 61 (10.1) | 54 (8.9) | 49 (8.1) | 58 (9.6) | |
| No | 2198 (90.8) | 544 (89.9) | 550 (91.1) | 555 (91.9) | 549 (90.4) | |
| Stroke, n (%) | 0.120 | |||||
| Yes | 161 (6.7) | 52 (8.6) | 41 (6.8) | 33 (5.5) | 35 (5.8) | |
| No | 2259 (93.3) | 553 (91.4) | 563 (93.2) | 571 (94.5) | 572 (94.2) | |
| Energy (kcal) | 1854.4 ± 807.1 | 1208.7 ± 487.6 | 1643.3 ± 495.0 | 2007.3 ± 577.1 | 2556.0 ± 900.7 | < 0.001 |
| Zinc(mg) | 9.9 ± 5.9 | 5.9 ± 3.0 | 8.6 ± 3.7 | 10.5 ± 4.7 | 14.7 ± 7.1 | < 0.001 |
| Iron(mg) | 13.9 ± 8.1 | 8.3 ± 4.8 | 12.2 ± 5.6 | 14.8 ± 6.6 | 20.2 ± 9.4 | < 0.001 |
| Selenium (mcg) | 102.5 ± 60.1 | 66.9 ± 32.9 | 90.0 ± 38.3 | 111.7 ± 44.8 | 141.2 ± 82.9 | < 0.001 |
| Fat(gm) | 71.6 ± 40.3 | 45.2 ± 24.1 | 63.0 ± 28.2 | 76.5 ± 31.1 | 101.4 ± 49.5 | < 0.001 |
| Saturated fatty acids (gm) | 22.6 ± 13.8 | 15.1 ± 8.9 | 20.5 ± 10.7 | 24.3 ± 11.8 | 30.5 ± 17.5 | < 0.001 |
| Cognitive score | ||||||
| DSST | 46.7 ± 17.0 | 42.6 ± 17.3 | 45.7 ± 16.8 | 48.0 ± 16.8 | 50.5 ± 16.1 | < 0.001 |
| AFT | 16.8 ± 5.5 | 15.5 ± 5.3 | 16.6 ± 5.5 | 17.2 ± 5.4 | 18.1 ± 5.6 | < 0.001 |
| CERAD-IRT | 19.1 ± 4.6 | 18.6 ± 4.6 | 19.0 ± 4.5 | 19.2 ± 4.5 | 19.6 ± 4.6 | 0.003 |
| CERAD-DRT | 6.0 ± 2.3 | 5.8 ± 2.3 | 5.9 ± 2.3 | 6.0 ± 2.3 | 6.3 ± 2.2 | < 0.001 |
| Z score | 0.00 ± 0.78 | − 0.17 ± 0.81 | − 0.04 ± 0.77 | 0.04 ± 0.77 | 0.17 ± 0.75 | < 0.001 |
%, weighted proportion; BMI body mass index; DSST Digit Symbol substation test; AFT Animal Fluency Test; CERAD Consortium to Establish a Registry for Alzheimer’s disease; CERAD-IRT immediate recall in CERAD trial; CERAD-DRT, delayed recall in CERAD trial; Z score is average of the standardized scores of DSST, AFT, CERAD-IRT, CERAD-DRT; SD standard deviation; IQR interquartile range; Q1–Q4 Quartile according to copper intake.
Relationship between dietary copper intake and cognitive function
Table 2 presents the results of the multivariate linear regression analysis of the association between copper consumption and cognitive function. Our research indicated a positive association between copper intake and cognitive scores. In the crude model, dietary copper intake expressed as a continuous variable (per 1 mg/day), showed an increased association with DSST (β = 1.08, 95% CI 0.55,1.60), AST (β = 0.44, 95% CI 0.27,0.61), CERAD-IRT (β = 0.20, 95% CI 0.05,0.34), CERAD-DRT (β = 0.08, 95% CI 0.01,0.16) and Z score (β = 0.06, 95% CI 0.03,0.08). These results remained consistent when dietary copper intake was analyzed as a categorical variable. Participants in the highest quartile (Q4) exhibited higher cognition scores (DSST: β = 7.90, 95% CI 6.01,9.78; AFT: β = 2.55, 95% CI 1.94,3.16; CERAD-IRT: β = 0.96, 95% CI 0.45,1.48; CERAD-DRT: β = 0.54, 95% CI 0.29,0.80; Z score: β = 0.34, 95% CI 0.26,0.43) compared to those in the lowest quartile (Q1). Moreover, as copper intake increased, the cognitive scores also showed a gradual increase (p for trend < 0.001). These associations remained statistically significant across all multivariate linear regression models, even after adjusting for various covariates such as age, gender, race/ethnicity, poverty income ratio (PIR), marital status, education level, body mass index (BMI), smoking status, alcohol status, hypertension, and diabetes mellitus, cardiovascular disease history, stroke, dietary energy, zinc, iron, selenium, fat, and total saturated fatty acids. In Model 4, compared with Q1, participants in Q4 groups were found with higher DSST (β = 3.80, 95% CI 1.90,5.70), AFT (β = 1.23, 95% CI 0.48,1.99), CERAD-IRT (β = 0.58, 95% CI − 0.06,1.22), CERAD-DRT (β = 0.47, 95% CI 0.15,0.80), and Z score (β = 0.20, 95% CI 0.10,0.29.The relationships between the covariates and cognitive function are shown in supplementary Table 1.
Table 2.
Multivariable linear regression to assess the association of copper intake with cognitive function.
| Variable | DSST | AFT | CERAD-IRT | CERAD-DRT | Z score | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| β (95% CI) | p-value | β (95% CI) | p-value | β (95% CI) | p-value | β (95% CI) | p-value | β (95% CI) | p-value | |
| Model 1 | Model 1 | Model 1 | Model 1 | Model 1 | ||||||
| Copper (mg/day) |
1.08 (0.55, 1.60) |
< 0.001 |
0.44 (0.27, 0.61) |
< 0.001 |
0.20 (0.05, 0.34) |
0.007 |
0.08 (0.01, 0.16) |
0.020 |
0.06 (0.03, 0.08) |
< 0.001 |
| Copper quartile | ||||||||||
| Q1(< 0.76) | 0(reference) | 0(reference) | 0(reference) | 0(reference) | 0(reference) | |||||
| Q2(0.76–1.04) |
3.06 (1.17, 4.95) |
0.002 |
1.09 (0.48, 1.70) |
< 0.001 |
0.41 (− 0.11, 0.93) |
0.119 |
0.10 (− 0.16, 0.35) |
0.465 |
0.13 (0.04, 0.21) |
0.004 |
| Q3(1.05–1.43) |
5.41 (3.53, 7.30) |
< 0.001 |
1.67 (1.05, 2.28) |
< 0.001 |
0.55 (0.04, 1.07) |
0.035 |
0.19 (− 0.07, 0.45) |
0.151 |
0.21 (0.12, 0.29) |
< 0.001 |
| Q4(≥ 1.44) |
7.90 (6.01, 9.78) |
< 0.001 |
2.55 (1.94, 3.16) |
< 0.001 |
0.96 (0.45, 1.48) |
< 0.001 |
0.54 (0.29, 0.80) |
< 0.001 |
0.34 (0.26, 0.43) |
< 0.001 |
| p for trend | < 0.001 | < 0.001 | < 0.001 | < 0.001 | < 0.001 | |||||
| Model 2 | Model 2 | Model 2 | Model 2 | Model 2 | ||||||
| Copper (mg/day) |
0.41 (0.02, 0.80) |
0.04 |
0.24 (0.09, 0.39) |
0.002 |
0.15 (0.02, 0.28) |
0.027 |
0.06 (0.00, 0.13) |
0.063 |
0.03 (0.01, 0.05) |
0.001 |
| Copper quartile | ||||||||||
| Q1(< 0.76) | 0(reference) | 0(reference) | 0(reference) | 0(reference) | 0(reference) | |||||
| Q2(0.76–1.04) |
1.25 (− 0.16, 2.66) |
0.082 |
0.68 (0.12, 1.23) |
0.017 |
0.27 (− 0.20, 0.74) |
0.264 |
0.04 (− 0.20, 0.28) |
0.735 |
0.07 (0.00, 0.14) |
0.057 |
| Q3(1.05–1.43) |
2.68 (1.25, 4.10) |
< 0.001 |
0.95 (0.39, 1.50) |
0.001 |
0.37 (− 0.10, 0.85) |
0.126 |
0.09 (− 0.14, 0.33) |
0.438 |
0.11 (0.04, 0.18) |
0.002 |
| Q4(≥ 1.44) |
3.74 (2.30, 5.19) |
< 0.001 |
1.41 (0.85, 1.98) |
< 0.001 |
0.64 (0.16, 1.12) |
0.009 |
0.38 (0.14, 0.62) |
0.002 |
0.20 (0.12, 0.27) |
< 0.001 |
| p for trend | < 0.001 | < 0.001 | 0.010 | 0.003 | < 0.001 | |||||
| Model 3 | Model 3 | Model 3 | Model 3 | Model 3 | ||||||
| Copper (mg/day) |
0.36 (− 0.02, 0.75) |
0.064 |
0.23 (0.08, 0.39) |
0.003 |
0.14 (0.01, 0.27) |
0.033 |
0.06 (0.00, 0.13) |
0.068 |
0.03 (0.01, 0.05) |
0.002 |
| Copper quartile | ||||||||||
| Q1(< 0.76) | 0(reference) | 0(reference) | 0(reference) | 0(reference) | 0(reference) | |||||
| Q2(0.76–1.04) |
1.03 (− 0.37, 2.42) |
0.149 |
0.65 (0.10, 1.20) |
0.021 |
0.24 (− 0.23, 0.71) |
0.311 |
0.04 (− 0.20, 0.27) |
0.772 |
0.06 (− 0.01, 0.13) |
0.084 |
| Q3(1.05–1.43) |
2.30 (0.89, 3.71) |
0.001 |
0.88 (0.33, 1.44) |
0.002 |
0.32 (− 0.16, 0.80) |
0.189 |
0.08 (− 0.16, 0.32) |
0.536 |
0.10 (0.03, 0.17) |
0.006 |
| Q4(≥ 1.44) |
3.20 (1.76, 4.64) |
< 0.001 |
1.34 (0.77, 1.91) |
< 0.001 |
0.58 (0.09, 1.06) |
0.020 |
0.36 (0.11, 0.61) |
0.004 |
0.18 (0.11, 0.25) |
< 0.001 |
| p for trend | < 0.001 | < 0.001 | 0.022 | 0.005 | < 0.001 | |||||
| Model 4 | Model 4 | Model 4 | Model 4 | Model 4 | ||||||
| Copper (mg/day) |
0.22 (− 0.21, 0.64) |
0.316 |
0.16 (− 0.01, 0.32) |
0.066 |
0.12 (− 0.02, 0.26) |
0.095 |
0.05 (− 0.02, 0.13) |
0.139 | 0.02 (0.00, 0.04) | 0.035 |
| Copper quartile | ||||||||||
| Q1(< 0.76) | 0(reference) | 0(reference) | 0(reference) | 0(reference) | 0(reference) | |||||
| Q2(0.76–1.04) |
1.25 (− 0.20, 2.69) |
0.091 |
0.62 (0.05, 1.19) |
0.033 |
0.24 (− 0.24, 0.73) |
0.328 |
0.07 (− 0.17, 0.32) |
0.565 |
0.07 (0.00, 0.14) |
0.067 |
| Q3(1.05–1.43) |
2.72 (1.15, 4.30) |
0.001 |
0.84 (0.22, 1.46) |
0.008 |
0.34 (− 0.19, 0.87) |
0.211 |
0.15 (− 0.12, 0.42) |
0.277 |
0.11 (0.03, 0.19) |
0.005 |
| Q4(≥ 1.44) |
3.80 (1.90, 5.70) |
< 0.001 |
1.23 (0.48, 1.99) |
0.001 |
0.58 (− 0.06, 1.22) |
0.078 |
0.47 (0.15, 0.80) |
0.004 |
0.20 (0.10–0.29) |
< 0.001 |
| p for trend | < 0.001 | 0.001 | 0.084 | 0.006 | < 0.001 | |||||
CI confidence interval; Q1–Q4 Quartile according to copper intake; DSST Digit Symbol substation test AFT Animal Fluency Test; CERAD Consortium to Establish a Registry for Alzheimer’s disease; CERAD-IRT immediate recall in CERAD trial; CERAD-DRT delayed recall in CERAD trial; Z score is average of the standardized scores of DSST-AFT-CERAD-IRT-CERAD-DRT; Model 1 was the crude model without adjustment for covariates. Model 2 was adjusted for age- gender- race/ethnicity- poverty income ratio-marital status-and education level. Model 3 was adjusted as for Model 2- additionally adjusted for body mass index-smoking status-alcohol status-hypertension-diabetes mellitus-cardiovascular disease history and stroke. Model 4 was adjusted as for Model 3 additionally adjusted for energy-zinc-iron-selenium-fat-and total saturated fatty acids.
Non-linear relationships
According to the fully adjusted model, dietary copper intake did not show a nonlinear relationship with CERAD-IRT and CERAD-DRT (nonlinear, p > 0.05), while the association with DSST, AFT, and Z scores showed an inverse L-shaped curve in the restricted cubic spline (nonlinear, p < 0.05). (Fig. 2). Threshold effect analysis identified an inflection point of 1.63 mg/day for DSST, 1.42 mg/day for AFT, and 1.22 mg/day for the Z score. A positive link was discovered between copper intake and DSST (β = 4.16, 95% CI = 1.92,6.40), AFT (β = 1.19, 95% CI 0.15,2.24), Z score (β = 0.16, 95% CI 0.01,0.36) before the turning point. Nevertheless, the relationship between dietary copper intake and DSST (β = − 0.14, 95% CI − 2.95,2.67), AFT (β = -0.43, 95% CI − 1.40,0.54), Z score (β = 0.07, 95% CI − 0.03,0.16) lost their statistical significance beyond the inflection point. This indicates that the cognitive scores no longer increase with increasing dietary copper intake. (Table 3).
Fig. 2.
Association between dietary copper intake and cognitive performance in CERAD-IRT, CERAD-DRT, DSST, AFT and Z score. Solid and dashed lines represent the predicted value and 95% confidence intervals. Adjusted for age, gender, race/ethnicity, poverty income ratio, marital status, education level, body mass index, smoking status, alcohol status, hypertension, diabetes mellitus, cardiovascular disease history, stroke, energy, zinc, iron, selenium, fat, and total saturated fatty acids. Only 99% of the data is shown. DSST, Digit Symbol substation test AFT, Animal Fluency Test; CERAD, Consortium to Establish a Registry for Alzheimer’s disease; CERAD-IRT, immediate recall in CERAD trial; CERAD-DRT, delayed recall in CERAD trial; Z score is average of the standardized scores of DSST, AFT, CERAD-IRT, CERAD-DRT.
Table 3.
Threshold effect analysis of the relationship of dietary copper intake and cognitive function.
| Copper intake (mg/day) | Adjusted β (95%CI) | p-value |
|---|---|---|
| DSST | ||
| < 1.63 | 4.16(1.92, 6.40) | < 0.001 |
| ≥ 1.63 | − 0.14 (− 2.95, 2.67) | 0.922 |
| Likelihood ratio test | 0.001 | |
| AFT | ||
| < 1.42 | 1.19(0.15, 2.24) | 0.025 |
| ≥ 1.42 | − 0.43(− 1.40, 0.54) | 0.384 |
| Likelihood ratio test | 0.007 | |
| Z score | ||
| < 1.22 | 0.16 (0.01, 0.36) | 0.034 |
| ≥ 1.22 | 0.07 (− 0.03, 0.16) | 0.167 |
| Likelihood Ratio test | 0.008 |
CI confidence interval; DSST Digit Symbol substation test AFT Animal Fluency Test; Z score is average of the standardized scores of DSST-AFT-CERAD-IRT-CERAD-DRT; Adjusted for age, gender, race/ethnicity, poverty income ratio, marital status, education level, body mass index, smoking status, alcohol status, hypertension, diabetes mellitus, cardiovascular disease history, stroke, energy, zinc, iron, selenium, fat, and total saturated fatty acids. Only 99% of the data is shown.
Subgroup analysis
The study also evaluated potential modifiers of the association between dietary copper intake and cognitive scores, including gender (male or female), age (60–69, 70–79, or ≥ 80 years), BMI (< 25, 25–30, or > 30 kg/m2), hypertension (yes or, no), diabetes mellitus (yes or, no), cardiovascular disease (yes or, no) and stroke (yes or, no). In the participants with a history of stroke, copper intake was associated with an increased Z score. Specifically, individuals in Q4 exhibited a higher Z score compared to those in Q1 (β = 0.55, 95%CI = 0.24,0.86; p for interaction = 0.009). No significant interactions were found between dietary copper intake and participant characteristics or other potential modifier (p for interaction > 0.05). (Fig. 3).
Fig. 3.
Stratified analyses of the association between cognitive performance and dietary copper intake according to baseline characteristics in Z score test. Note: The P value for interaction represents the likelihood of interaction between the variable and copper. CI confidence interval.
Discussion
This cross-sectional study found that copper intake was associated with better cognitive function test scores in American older adults. This association remained statistically significant even after controlling for all potential confounding variables. Additionally, we observed a reverse L-shaped relationship between dietary copper intake and cognitive scores (including DSST, AFT, and Z scores) with significant non-linearity. The optimal reference thresholds for copper intake were 1.63 mg/day for the DSST, 1.42 mg/day for the AFT, and 1.22 mg/day for the Z score. Below these inflection points, an increased copper intake was associated with improved cognitive scores. However, when copper intake surpassed these thresholds, the correlation between dietary copper intake and cognitive performance lost statistical significance, indicating that cognitive scores do not continue to increase with further increases in copper consumption in older adults. This aligns with increasing evidence that copper significantly affects impacts brain health and disease12,13.
Copper is a vital trace element that plays a critical role in numerous physiological processes, including neurotransmitter synthesis, cellular energy production, and antioxidant defenses14. It serves as an essential cofactor for several enzymes necessary for optimal functioning of the nervous system15. Research indicates that disruptions in copper homeostasis can result in neurological dysfunction and contribute to multiple neurodegenerative disorders such as Wilson’s disease and AD16,17. The relationship between copper and cognitive function is intricate, with both deficiencies and excesses potentially leading to neurological disorders. A prospective cohort study involving Chinese older adults demonstrated a negative and non-linear relationship between dietary copper intake and cognitive decline, identifying an inflection point at approximately 1.3 mg/day of dietary copper intake. Similarly, Wang et al. found a non-linear association between copper intake and cognitive functioning. These results are consistent with our findings18,19. However, Zhao et al. reported the opposite associations, which may be due to differences in cognitive and dietary intake assessment methods20.
Subgroup analyses demonstrated a significant interaction between the Z score and stroke, indicating that an elevated dietary copper intake was associated with a more pronounced increase in Z score among participants with a history of stroke. Several studies have reported a relationship between copper levels and stroke risk. A study utilizing data from 2013 to 2018 NHANES suggested that an increased dietary copper intake was associated with a reduced risk of stroke. The prevalence of stroke decreased with an increase in dietary copper intake and showed a non-linear relationship21. Huuskonen et al. found that copper-targeted delivery attenuates neuronal damage in ischemic stroke22,23. Following a stroke, oxidative stress exacerbates neuronal damage. Copper, as a cofactor for Cu/Zn-SOD, helps reduce the production of reactive oxygen species. When serum copper levels are between 90 and 110 μg/dL (approximately 1.5 mg/day intake), SOD1 activity reaches its optimal state, reducing lipid peroxidation in the brain by 40%. This aligns with the DSST, AFT, and Z-score thresholds (1.2–1.6 mg/day) we observed. Copper influences microglial polarization after stroke. Low to moderate copper levels can inhibit NLRP3 inflammasome formation, causing microglia to transition from pro-inflammatory M1 type to anti-inflammatory M2 type, thereby reducing TNF-α and IL-1β levels while increasing IL-10 levels, thus reducing neuroinflammation. Additionally, copper-dependent enzymes are crucial for post-stroke repair: lysyl oxidase enhances vascular membrane integrity and promotes angiogenesis, while the HOCl-β-catenin axis activates the Wnt signalling pathway via copper-dependent myeloperoxidase, promoting neurogenesis in the hippocampus24,25.
Although the precise mechanisms underlying the relationship between dietary copper intake and cognitive function remain unknown, our findings are considered biologically plausible. Dietary copper is crucial for brain health and may confer protective effects on cognitive function through its involvement in antioxidant defense, neurotransmitter synthesis, and energy metabolism. Previous studies have shown that copper imbalance may be associated with pathological changes in neurodegenerative diseases, such as AD26. In a community-based study, researchers found a significant association between copper levels in the brain cognitive decline and the pathological burden of AD. Specifically, higher brain copper levels were associated with slower cognitive decline and fewer features of AD pathology12. The role of copper in the brain may be related to its function in neurotransmission, antioxidant defense and energy metabolism. Copper is a cofactor for a variety of enzymes that play a key role in the normal function of neurons27. For example, copper is involved in the synthesis of superoxide dismutase, which is essential for scavenging reactive oxygen species from cells, thereby protecting neurons from oxidative stress28. One of the major contributors to neuronal damage and cognitive decline29.In addition, copper may affect cognitive function by regulating the synthesis and release of neurotransmitters30. Studies have shown that copper plays an important role in the synthesis of acetylcholine, a neurotransmitter closely related to learning and memory11. Notably, the effects of copper on cognitive function may be influenced by other dietary factors such as fat and saturated fatty acid intake. Wang et al. suggested that the relationship between copper intake and cognitive function could be moderated by these factors, highlighting the complexity of dietary interactions in cognitive health19.
We used a multidimensional, comprehensive Z score to assess cognitive function. In addition, dose–response and subgroup analyses were performed to determine the potential relationship between dietary copper intake and cognitive function. However, this study had some limitations that must be considered. First, due to the cross-sectional nature of this study, we were unable to determine the temporal association between dietary copper intake and cognitive function. Further randomized controlled trials are required to confirm these findings. Second, despite comprehensive adjustment for demographic, lifestyle, and clinical covariates, residual confounding remains a major concern, especially for dietary studies. Nutrient intakes (like copper) are highly correlated with other dietary components (e.g., fats, vitamins, phytochemicals) and overall dietary patterns, which may themselves influence cognitive health. Although we adjusted for total energy and zinc, iron, selenium, fat, and total saturated fatty acid intake, unmeasured or imperfectly measured confounders (e.g., other bioactive compounds in copper-rich foods, dietary interactions) could still bias our estimates. Third, because the copper dietary data were collected from self-reported 24-h dietary reviews, some measurement and information biases may exist. Finally, additional research is needed to determine whether the current findings can be extrapolated to other populations based on this study of older adults in the US.
Conclusions
In conclusion, the current study indicates a potential association between dietary copper intake and enhanced cognitive function in American older adults, particularly among those with a history of stroke. Dose–response analysis suggested an optimal copper intake level, with an inflection point of approximately1.22 mg per day. However, further longitudinal studies are necessary to confirm these findings.
Materials and methods
Data sources and study population
This cross-sectional study was conducted using NHANES data from 2011 to 2014, carried out by the National Center for Health Statistics (NCHS)31. The NHANES program employs a stratified, multistage probability cluster sampling design to conduct biennial surveys and ensure that the sampled population accurately represents the entire U.S. population. Health interviews were conducted to collect sociodemographic information including age, gender, household income, and education levels, as well as health-related lifestyle variables such as alcohol consumption, and smoking habits. All procedures were approved by the NCHS Research Ethics Committee, and the participants provided written informed consent32.
In this study, participants with incomplete cognitive information and those with incomplete copper intake information were excluded. The final sample consisted of 2420 individuals aged ≥ 60 years. We followed the STROBE guidelines for reporting observational studies in this study33.
Measures
Dietary copper intake
The NHANES utilized a 24-h dietary recall questionnaire, administered to all participants to gather detailed information on the types and quantities of food consumed in the preceding 24 h. All participants were eligible to participate in two 24-h dietary recall interviews, with the collected data used to determine each individual’s daily copper intake. The first dietary recall interview was conducted in person at the Mobile Examination Center (MEC), while the second interview was conducted via telephone within a 3 to 10-day interval. The copper intake of the participants was determined by averaging the two 24-h dietary recalls. Participants were categorized based on their copper intake34.
Assessment of cognitive function
In the NHANES study, participants underwent a series of cognitive function assessments to evaluate their memory and executive abilities. The immediate and delayed verbal list learning tests (CERAD-IRT and CERAD-DRT), administered by the Alzheimer’s Disease Word Learning Registry Consortium, are designed to assess the capacity for novel linguistic acquisition35. Participants were instructed to read aloud a list of 10 unrelated words and subsequently perform an immediate recall of as many words as possible. The scores from the initial three tests were aggregated for the analysis. Additionally, a delayed recall test was conducted approximately 8–10 min after the start of the word learning trial, with a score range of 0–10. In the Animal Fluency Test (AFT), participants named as many animals in one minute as possible to measure their verbal and executive abilities36. One point was awarded for each animal name. The Digit Symbol Substitution Test (DSST) is a time-limited evaluation designed to measure processing speed and executive function by requiring participants to transcribe symbols corresponding to digits, based on a provided key37. Participants were given two minutes to match symbols to numbers within 133 boxes, using the keys provided at the top. The scores were determined based on the number of correct matches, with possible scores ranging from 0 to 133.
Standard scores for all the cognitive tests, including the DSST, AFT, CERAD-IRT, and CERAD-DRT, were calculated using means and standard deviations (SDs). The global cognitive Z score was determined by averaging the standardized scores of the four tests. In all the tests, higher scores indicated superior cognitive performance9,38.
Covariates
Multiple potential covariates were assessed based on the published research and clinical judgments including gender, age, BMI, race, marital status, education, PIR, smoking status, alcohol status, dietary energy, zinc, iron, selenium, fat, and total saturated fatty acids, and a history of hypertension, diabetes mellitus, cardiovascular disease, and stroke9,39,40. The classification of race and ethnicity included non-Hispanic White, non-Hispanic Black, Mexican American, and other ethnic groups. Marital status was categorized as married or living with a partner, and educational attainment was divided into three levels: < 9 years, 9–12 years, and > 12 years of education. The PIR was employed to stratify family income into three categories: low (PIR ≤ 1.3), medium (PIR > 1.3 to 3.5), and high (PIR > 3.5). Smoking status was classified as never smokers (smoked fewer than 100 cigarettes), current smokers, and former smokers (quit smoking after consuming more than 100 cigarettes). Alcohol status was also divided into three categories: never drinking (consumed less than 12 alcoholic drinks in a lifetime), current drinking, and former drinking (consumed at least 12 drinks in a lifetime).
Statistical analysis
Continuous variables are presented as means and SD, whereas categorical variables are presented as percentages and proportions. Chi-square test and one-way analysis of variance were used to compare the baseline characteristics of the study participants.
Dietary copper intake was categorized into quartiles, with Q1 as the reference group. Multivariate linear regression explored copper intake as both a continuous and categorical variable. β and 95% CI were calculated to assess the relationship between copper intake and cognitive function. Model 1 represents an unadjusted model without controlling for any variables. Model 2 was adjusted for age, gender, race/ethnicity, PIR, marital status, and education level. Model 3 was adjusted for Model 2, and further adjusted for BMI, smoking status, alcohol status, hypertension, diabetes mellitus, history of cardiovascular disease, and stroke. Model 4 was adjusted for Model 3, and further adjusted for dietary energy, zinc, iron, selenium, fat, and total saturated fatty acids. Linear trend tests were performed by treating categorical variables as continuous variables. To account for the non-linear relationship between copper intake and cognitive test scores, we used a generalized additive model and smooth curve fitting to address nonlinearity. In addition, a two-piecewise binary linear regression model was used to explain the nonlinearity further. The likelihood-ratio test and bootstrap resampling method were used to determine inflection points.
Subgroup analysis examined the relationship between copper intake and cognitive function according to gender (female or male), age (60–69, 70–79, or ≥ 80 years), BMI (< 25, 25–30, or > 30 kg/m2), hypertension (yes or no), diabetes (yes or no), coronary heart disease (yes or no), and stroke (yes or no). A test for interaction in the linear regression model was used to compare β and 95% CI between the analyzed subgroups.
All analyses were performed using statistical software packages Free Statistics software version 2.0. A two-sided p < 0.05 was considered statistically significant.
Supplementary Information
Acknowledgements
Thank The Fourth Hospital of Hebei Medical University for their assistance with manuscript submission.
Abbreviations
- NHANES
National Health and Nutrition Examination Survey
- NCHS
National Center for Health Statistics
- MEC
Mobile Examination Center
- DSST
Digit symbol substitution test
- AFT
Animal fluency test
- CERAD
Consortium to establish a registry for Alzheimer’s disease
- CERAD-IRT
Immediate verbal list learning
- CERAD-DRT
Delayed verbal list learning
- MCI
Mild cognitive impairment AD Alzheimer’s disease
- PIR
Poverty income ratio
- Q
Quartile
- SD
Standard deviation
- IQR
Median and interquartile range
- CI
Confidence intervals
Author contributions
Conceived and designed, WAJ; formal analysis, WAJ and FFY; methodology, WAJ and JPS; writing—original draft, WAJ and KSZ; writing review and editing, WAJ and FFY. All authors contributed to the article and approved the submitted version the study.
Data availability
The National Health and Nutrition Examination Survey data are publicly available at https://wwwn.cdc.gov/nchs/nhanes which is publicly available. The data underlying this article will be shared on reasonable request with the corresponding author.
Declarations
Competing interests
The authors declare no competing interests.
Ethics approval
The National Health and Nutrition Examination Survey (NHANES) is conducted by the Centers for Disease Control and Prevention (CDC) and the National Center for Health Statistics (NCHS). The NCHS Research Ethics Review Committee reviewed and approved the NHANES study protocol. All participants signed written informed consent.
Consent for publication
The authors have no ethical, legal, and financial conflicts related to the article. All authors read and approved the manuscript for publication.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-025-09280-9.
References
- 1.Chapman, S. et al. Association of subjective cognitive decline with progression to dementia in a cognitively unimpaired multiracial community sample. Neurology100, e1020–e1027 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.GBD 2021 Nervous system disorders collaborators. Global, regional, and national burden of disorders affecting the nervous system, 1990–2021: A systematic analysis for the global burden of disease study 2021. Lancet Neurol.23, 344–381 (2024). [DOI] [PMC free article] [PubMed]
- 3.Manchia, M., Paribello, P., Pinna, M. & Faa, G. The role of copper overload in modulating neuropsychiatric symptoms. Int. J. Mol. Sci.25, 6487 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.GBD 2019 Dementia forecasting collaborators. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the global burden of disease study 2019. Lancet Public Health 7, e105-e125 (2022). [DOI] [PMC free article] [PubMed]
- 5.Zhang, X.-X. et al. The epidemiology of Alzheimer’s disease modifiable risk factors and prevention. J. Prev. Alzheimers Dis.8, 313–321 (2021). [DOI] [PubMed] [Google Scholar]
- 6.Guarnieri, L. et al. Impact of micronutrients and nutraceuticals on cognitive function and performance in Alzheimer’s disease. Ageing Res. Rev.95, 102210 (2024). [DOI] [PubMed] [Google Scholar]
- 7.Fekete, M. et al. Improving cognitive function with nutritional supplements in aging: A comprehensive narrative review of clinical studies investigating the effects of vitamins, minerals, antioxidants, and other dietary supplements. Nutrients15, 5116 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Y, W., X, S., H, M., X, Q. & H, W. Iron, zinc and copper from cereal food sources and cognitive performance in older adults in china. Iran. J. Public Health50, (2021). [DOI] [PMC free article] [PubMed]
- 9.Jia, W. et al. Association between dietary vitamin B1 intake and cognitive function among older adults: A cross-sectional study. J. Transl. Med.22, 165 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Chen, J. et al. The molecular mechanisms of copper metabolism and its roles in human diseases. Pflug Arch. Eur. J. Phy.472, 1415–1429 (2020). [DOI] [PubMed] [Google Scholar]
- 11.Zhong, G. et al. Insights into the role of copper in neurodegenerative diseases and the therapeutic potential of natural compounds. Curr. Neuropharmacol.22, 1650–1671 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Agarwal, P. et al. Brain copper may protect from cognitive decline and alzheimer’s disease pathology: A community-based study. Mol. Psychiatr.27, 4307–4313 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bost, M. et al. Dietary copper and human health: Current evidence and unresolved issues. J. Trace Elem. Med. Biol.35, 107–115 (2016). [DOI] [PubMed] [Google Scholar]
- 14.Baker, Z. N., Cobine, P. A. & Leary, S. C. The mitochondrion: A central architect of copper homeostasis. Metallomics9, 1501–1512 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wang, J. et al. Copper supplementation alleviates hypoxia-induced ferroptosis and oxidative stress in neuronal cells. Int. J. Mol. Med.54, 117 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Członkowska, A. et al. Wilson disease. Nat. Rev. Dis. Prim.4, 21 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sensi, S. L., Granzotto, A., Siotto, M. & Squitti, R. Copper and zinc dysregulation in Alzheimer’s disease. Trends Pharmacol. Sci.39, 1049–1063 (2018). [DOI] [PubMed] [Google Scholar]
- 18.Meng, Q. et al. Association between dietary copper intake and cognitive decline: A perspective cohort study in Chinese elderly. Am. J. Geriatr. Psychiatry31, 753–763 (2023). [DOI] [PubMed] [Google Scholar]
- 19.Wang, X. et al. Threshold effects of total copper intake on cognitive function in US older adults and the moderating effect of fat and saturated fatty acid intake. J. Acad. Nutr. Diet.121, 2429–2442 (2021). [DOI] [PubMed] [Google Scholar]
- 20.Zhao, D. et al. Dietary intake levels of iron, copper, zinc, and manganese in relation to cognitive function: A cross-sectional study. Nutrients15, 704 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.L, Y., X, C., H, C. & L, Z. Dietary copper intake and risk of trokse in daults: a case-control study based on National Health and Nutrition Examination Survey 2013–2018. Nutrients14, (2022). [DOI] [PMC free article] [PubMed]
- 22.Huuskonen, M. T. et al. The copper bis(thiosemicarbazone) complex cuII(atsm) is protective against cerebral ischemia through odulmation of the inflammatory milieu. Neurotherapeutics14, 519–532 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Peng, G., Huang, Y., Xie, G. & Tang, J. Exploring copper’s role in stroke: Progress and treatment approaches. Front. Pharmacol.15, 1409317 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Li, L., et al. From copper homeostasis to cuproptosis: a new perspective on CNS immune regulation and neurodegenerative diseases. Front. Neurol.29,16:1581045(2025). [DOI] [PMC free article] [PubMed]
- 25.Ban, X. X. et al. Copper metabolism and cuproptosis: Molecular mechanisms and therapeutic perspectives in neurodegenerative diseases. Curr. Med. Sci.44(1), 28–50 (2024). [DOI] [PubMed] [Google Scholar]
- 26.Mezzaroba, L., Alfieri, D. F., Colado Simão, A. N. & Vissoci Reiche, E. M. The role of zinc, copper, manganese and iron in neurodegenerative diseases. Neurotoxicology74, 230–241 (2019). [DOI] [PubMed]
- 27.Maung, M. T. et al. The molecular and cellular basis of copper dysregulation and its relationship with human pathologies. FASEB J.35, e21810 (2021). [DOI] [PubMed] [Google Scholar]
- 28.Zhang, Y. et al. Copper induces cognitive impairment in mice via modulation of cuproptosis and CREB Signaling. Nutrients15, 972 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Wang, L. et al. Effectiveness and mechanisms of combined use of antioxidant nutrients in protecting against oxidative stress-induced neuronal loss and related neurological deficits. CNS Neurosci. Ther.30, e14886 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bhattacharjee, A., Ghosh, S., Chatterji, A. & Chakraborty, K. Neuron-glia: Understanding cellular copper homeostasis, its cross-talk and their contribution towards neurodegenerative diseases. Metallomics12, 1897–1911 (2020). [DOI] [PubMed] [Google Scholar]
- 31.National Center for Health Statistics. NHANES Survey Methods and Analytic Guidelines. https://wwwn.cdc.gov/nchs/nhanes/analyticguidelines.aspx#analytic-guidelines.
- 32.CDC. Ethics Review Board Approval. National Health and Nutrition Examination Survey https://www.cdc.gov/nchs/nhanes/about/erb.html (2024).
- 33.Ghaferi, A. A., Schwartz, T. A. & Pawlik, T. M. STROBE reporting guidelines for observational studies. JAMA Surg.156, 577–578 (2021). [DOI] [PubMed] [Google Scholar]
- 34.National health and nutrition examination survey. Interviewer procedures manual (2022). https://stacks.cdc.gov/view/cdc/127210.
- 35.Lu, Z. et al. The Relationship between alternative healthy Ddet index and cognitive function in the older adults: the mediating effect of depressive symptoms. Nutrients14, 2856 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Long, L. S., Wolpaw, J. T. & Leung, J. M. Sensitivity and specificity of the animal fluency test for predicting postoperative delirium. Can. J. Anaesth.62, 603–608 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Jaeger, J. Digit symbol substitution test: The case for sensitivity over specificity in neuropsychological testing. J. Clin. Psychopharmacol.38, 513–519 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Wei, J., Wang, L., Kulshreshtha, A. & Xu, H. Adherence to life’s simple 7 and cognitive function among older adults: The National Health and Nutrition Examination Survey 2011 to 2014. J. Am. Heart Assoc.11, e022959 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Tang, H., Zhang, X., Luo, N., Huang, J. & Zhu, Y. Association of dietary live microbes and nondietary prebiotic/probiotic intake with cognitive function in older adults: evidence from NHANES. J. Gerontol.: Ser. A glad175 (2023) 10.1093/gerona/glad175. [DOI] [PubMed]
- 40.Zhong, Q., Sun, W., Qin, Y. & Xu, H. Association of dietary α-carotene and β-carotene intake with low cognitive performance in older adults: A cross-sectional study from the National Health and Nutrition Examination Survey. Nutrients15, 239 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The National Health and Nutrition Examination Survey data are publicly available at https://wwwn.cdc.gov/nchs/nhanes which is publicly available. The data underlying this article will be shared on reasonable request with the corresponding author.