Evaluating dietary quality and dietary inflammatory potential in cognitive impairment patients in China

Abstract

INTRODUCTION

Diet plays a critical role in the prevention and treatment of cognitive impairment, yet dietary intake in patients with cognitive impairment in China remains insufficiently studied.

METHODS

A 25‐item semi‐quantitative food frequency questionnaire was used in a national online survey in China. The Dietary Inflammatory Index was calculated, and principal component analysis was used to identify dietary patterns.

RESULTS

The average daily energy intake was 1131.59 ± 486.30 kcal, below recommended levels. Patients in the pro‐inflammatory diet group had lower intakes of nutrients than those in the anti‐inflammatory group. Ordinal logistic regression indicated that adherence to a high‐quality protein dietary pattern was associated with lower cognitive impairment severity, whereas the high cereal and snack dietary pattern was associated with increased severity.

CONCLUSION

Inadequate intake of essential nutrients, particularly energy and protein, exacerbates cognitive decline. Promoting high‐quality protein intake while reducing low‐quality carbohydrates may mitigate cognitive deterioration and improve patient health outcomes.

Highlights

• Cognitive impairment patients in China show significantly low intake of energy, protein, and essential nutrients, which contributes to malnutrition and exacerbates cognitive decline.

• The study finds a link between inadequate nutrient intake and increased dietary inflammatory potential, which may accelerate cognitive decline.

• The pro‐inflammatory deit with higher Dietary Inflammatory Index resulted from an overall insufficient intake of nutrients rather that the comsumption of excessively high levels of “pro‐inflammatory” foods in patients.

• Dietary patterns with higher high‐quality protein and lower low‐quality carbohydrate intake may benefit nutritional status and cognitive health in cognitively impaired patients.

1. INTRODUCTION

The disease burden and economic costs of cognitive impairment have increased significantly due to aging societies, making it a global public health priority as life expectancy rises in the world. ¹ , ² , ³ As is well known, cognitive impairment is associated with various physical and psychological issues, adversely affecting patients’ long‐term quality of life (QoL) and overall well‐being. ⁴ , ⁵ Patients with cognitive impairment frequently face nutritional challenges, including irregular meal consumption, difficulties with eating, and a decrease in appetite, primarily resulting from a decline in their functional abilities associated with aging. ⁵ , ⁶ Additionally, as the disease progresses, patients often experience behavioral and psychological disturbances, including apathy, disinhibition, agitation, depression, psychosis, changes in appetite, and sleep disorders. ⁶ , ⁷ These alterations also could lead to increased dependency, heightened energy demands, and difficulties with self‐feeding and swallowing, ultimately contributing to malnutrition and a decline in QoL. ⁸

Despite the growing understanding of the risk factors associated with dementia, there is still limited knowledge regarding effective management strategies for dietary and nutritional approaches. Recent studies have begun to explore the impact of various dietary patterns on cognitive health and dementia prevention, including the anti‐inflammatory diet, ⁹ , ¹⁰ , ¹¹ , ¹² the Dietary Approaches to Stop Hypertension (DASH) diet, the Mediterranean diet, and the Mediterranean–DASH Diet Intervention for Neurodegenerative Delay (MIND). ⁶ , ¹⁶ These diets share common characteristics, such as being anti‐inflammatory, plant based, and rich in poly‐ and monounsaturated fatty acids, while minimizing the intake of processed foods. Meanwhile, a series of studies has confirmed that synergistic interactions among the components of these diets play a crucial role in preventing or slowing cognitive decline. ¹⁷ , ¹⁸ , ¹⁹ Although there are numerous studies which have examined the association between diet and the risk of developing dementia, research specifically focusing on the dietary quality of patients with established cognitive impairment remains limited. Cognitive impairment patients are at a heightened risk of malnutrition, with literature reporting malnutrition prevalence rates of 32.52% and malnutrition risk rates of 46.80% among dementia patients. ⁶ Malnutrition can exacerbate the progression of cognitive decline, making it crucial to address this issue. Therefore, a thorough investigation into the dietary patterns and nutritional status of these patients is essential for developing targeted dietary interventions that may slow disease progression and enhance quality of life.

Here, by addressing the landscape of the dietary quality and specific nutritional challenges of patients with cognitive impairment, we aim to uncover critical factors affecting their overall well‐being and develop tailored nutritional intervention strategies that could help mitigate cognitive decline and improve their quality of life.

2. METHODS

2.1. Participants

This cross‐sectional study, conducted by Alzheimer's Disease China (ADC), the only official member of Alzheimer's Disease International (ADI) in China, initiated a nationwide dietary and nutritional survey for patients with cognitive impairment from July to September 2024. The survey encompassed all 30 provinces, autonomous regions, and municipalities across the country. The inclusion criteria for participants were as follows: (1) the patient or their caregiver was well informed about dietary intake over the past 1 year, (2) experiencing subjective cognitive decline or diagnosed with cognitive impairment by neurologists, (3) the patient or their caregiver was able to answer relevant questions, and (4) willingness to provide informed consent. Accordingly, we calculated the sample size for patients with cognitive impairment. Assuming a prevalence of cognitive impairment of 21.5%, with a margin of error of 3% and a confidence level of 95%, the minimum required sample size was determined to be 721. After increasing this by 20%, a total sample size of 866 participants was necessary.

RESRARCH IN CONTEXT

1. Systematic Review: We conducted a PubMed search for studies exploring dietary quality, dietary inflammatory potential, and cognitive impairment, with an emphasis on observational and interventional studies in older populations. While previous research underscores diet's role in cognitive health, few studies have focused specifically on dietary quality and inflammatory potential among patients with cognitive impairment.

2. Interpretation: This study found that patients with cognitive impairment generally have inadequate intake of energy, protein, and essential nutrients, increasing dietary inflammatory potential and potentially accelerating cognitive decline. Ordinal logistic regression analysis demonstrated that adherence to a high‐quality protein dietary pattern was significantly associated with lower severity of cognitive impairment, whereas a high cereal and snack dietary pattern was positively associated with more severe cognitive impairment. These findings underscore the suboptimal nutritional status of this population and suggest that overall dietary quality may play an important role in the clinical progression of cognitive impairment.

3. Future Directions: Our study highlights the importance of dietary interventions emphasizing high‐quality protein and reduced consumption of low‐quality carbohydrates for patients with cognitive impairment. Future research should investigate the long‐term impact of such dietary pattern on cognitive health and develop effective strategies to integrate this dietary approach into caregiver training and community support programs.

Based on the objectives of this study, we developed an online survey questionnaire that includes patients’ basic information and dietary intake (Supplementary Material in supporting information). Participants in this study were recruited from a population largely consistent with our previously published ADC study, ²⁰ with comparable inclusion criteria and a caregiver‐report approach. A pilot assessment was conducted to evaluate the feasibility and clarity of the questionnaire. Several practical issues were identified during the pilot phase. More than 80% of responses were completed by caregivers, and some respondents confused their own demographic information with that of the patient—particularly age and sex—resulting in inconsistencies in the data. In addition, the dietary recall section was relatively complex and time consuming, increasing respondent burden and the likelihood of minor reporting errors. In response to these concerns, demographic items such as patient age and sex were removed from the final version of the questionnaire to minimize confusion and improve data quality. Further, this focus on dietary intake is broadly consistent with previous dietary surveys in populations with chronic disease, in which the primary objective was to evaluate overall dietary patterns and nutritional challenges, rather than to perform stratified analyses based on demographic variables. ²¹ , ²² The last revision was consistent with the study's primary objective: to accurately assess dietary intake in individuals with cognitive impairment under real‐world conditions.

The questionnaire was accompanied by detailed instructions and informed consent. Additionally, a time limit was set for completing the questionnaire, and all questionnaires completed in < 5 minutes were excluded. Furthermore, consecutive questions with identical answers were also excluded. We initially recruited 1035 valid participants for this cross‐sectional study. However, 47 patients were excluded due to incomplete data, including missing dietary information. Additionally, 24 patients were excluded because of significant logical inconsistencies in their dietary data, such as reporting identical intake amounts for all food categories or abnormal energy intake. Consequently, the final sample used for analysis consisted of 964 patients with cognitive impairment. The study was approved by the human ethics research committee of Renji Hospital, Shanghai Jiao Tong University School of Medicine (KY2024‐119‐B). Informed consent was obtained from all participants before the study commenced.

2.2. Data collection of dietary intake

We collected basic information from patients or caregivers, including educational background, living conditions, and residential area. Additionally, specialized neurologists provided diagnoses such as subjective cognitive decline (SCD), mild cognitive impairment (MCI), and Alzheimer's disease (AD)–type dementia, and recorded diagnostic information and treatment plans that included apolipoprotein E (APOE) gene polymorphisms, pharmacological therapies, and lifestyle interventions.

We used the simplified 25‐item semi‐quantitative food frequency questionnaire designed by Gao to assess the dietary intake of the participants, ²³ detailed in Appendix A. This questionnaire has been validated for reliability and validity in the Chinese elderly population. The dietary survey was conducted through online responses, in which participants were asked about their average consumption amounts and frequencies of various foods over the past year. Consumption frequency was categorized into nine groups: never, less than monthly, one to three times monthly, one to two times weekly, three to four times weekly, five to six times weekly, once daily, twice daily, three or more times daily. The food intake amounts were categorized into six groups: ≤ 1 liang (50 g), 2 liang, 3 liang, 4 liang, ≥ 5 liang, and not applicable. The food categories included cereals, congee, flour‐based foods, desserts, fried foods, stuffed foods, whole grains, tubers, dairy products, eggs, red meat, poultry, processed meat products, freshwater fish, seafood, soy products, nuts, dark‐ green vegetables, light‐colored vegetables, edible mushrooms, fruits, sugary beverages, and alcoholic beverages. Each food item in the questionnaire was accompanied by images and weight labels to facilitate accurate reporting. Additionally, nutrient and total energy intake were calculated by multiplying the usual frequency and portion size of each food item by the nutrient content, using the food composition values from the China Food Composition Table (National Institute of Nutrition and Food Safety, Chinese Center for Disease Control and Prevention).

To evaluate dietary intake among individuals with cognitive impairment, we compared their food consumption patterns to two authoritative references: (1) The Chinese Dietary Guidelines for Older Adults (2022), ²⁴ issued by the National Health Commission of China and the Chinese Nutrition Society, which provides evidence‐based recommendations on nutrient intake and food group distribution for the older adults and (2) The China Nutrition and Health Surveillance (CNHS) 2015–2017, a large‐scale epidemiological dataset reflecting the dietary patterns of the general adults in China. ²⁵ , ²⁶ By incorporating both dietary recommendations and real‐world data, we aimed to assess the characteristics of dietary intake in individuals with cognitive impairment and explore potential deviations from established dietary guidelines and population‐based dietary patterns.

2.3. Dietary Inflammatory Index and dietary pattern

The Dietary Inflammatory Index (DII), developed by researchers at the University of South Carolina's Cancer Prevention and Control Program, is a tool designed to predict the inflammatory potential of an individual's diet. ²⁷ The DII is based on six inflammatory biomarkers: interleukin (IL)‐1β, IL‐4, IL‐6, IL‐10, tumor necrosis factor alpha (TNF‐α), and C‐reactive protein (CRP). Originally derived from the literature, the DII included 45 food items and nutrients. In this study, we calculated the DII using 28 food parameters, which included alcohol, vitamin B12, vitamin B6, β‐carotene, carbohydrates, cholesterol, energy, total fat, fiber, folic acid, iron, magnesium, monounsaturated fatty acids, niacin, n‐3 fatty acids, n‐6 fatty acids, protein, polyunsaturated fatty acids, riboflavin, saturated fat, selenium, thiamin, vitamin A, vitamin C, vitamin D, vitamin E, zinc, and isoflavones. To derive the final DII score, the individual's daily intake of each food parameter was first subtracted from the standard global mean and then divided by the standard deviation of that value to obtain z scores. These z scores were subsequently converted into percentile scores, doubled, and adjusted by subtracting 1 to minimize the impact of right skewness. The processed values were then multiplied by their corresponding inflammatory effect scores, as established through a literature review, and summed to calculate the overall DII score. A lower DII score reflects a more anti‐inflammatory diet, whereas a higher DII score indicates a more pro‐inflammatory diet. For the purposes of the present study, 25 food frequency questionnaire–derived food parameters were used for DII calculation.

To identify the primary dietary patterns based on food groups in this study, we used principal component analysis (PCA) as part of an exploratory factor analysis. In line with the Kaiser criterion, we retained factors with eigenvalues > 1. An orthogonal rotation (varimax rotation) was then applied to enhance the clarity of the factor structure and improve interpretability. The naming of the dietary patterns was based on the distinct characteristics of the food groups associated with each factor, supplemented by insights from previous research.

2.4. Statistical analysis

Descriptive statistical analysis, differential analysis, and regression analysis were carried out using SPSS version 23.0. Continuous variables are reported as mean ± standard deviation or median (interquartile range), while categorical variables are presented as percentages. One‐way analysis of variance (ANOVA) was used for normally distributed data, with Bonferroni post hoc tests applied for pairwise comparisons, the Kruskal–Wallis test for data with skewed distributions, and the chi‐squared test for categorical data.

The DII was categorized into four quartiles, referred to as Q1, Q2, Q3, and Q4. Q1 corresponds to the lowest quartile, representing the anti‐inflammatory dietary group, while Q4 corresponds to the highest quartile, indicating the pro‐inflammatory dietary group. Additionally, Spearman correlation analysis was used to assess the associations between DII scores and dietary patterns. Ordinal logistic regression models were used to assess the associations among DII, dietary patterns, and stage of cognitive impairment. Two models were constructed: Model 1 included DII or dietary pattern as the independent variable without covariate adjustment, while Model 2 additionally adjusted for APOE genotype and years of cognitive impairment diagnosis to account for potential genetic confounding. P for trend was calculated by treating DII and dietary pattern quartiles as continuous ordinal variables in the regression models, to evaluate potential dose–response relationships. A two‐sided P value < 0.05 was considered statistically significant.

3. RESULTS

3.1. Characteristics of the demographics of the study

In total, 964 patients with cognitive impairment were enrolled, including 321 (33.29%) with SCD, 150 (15.56%) with MCI, and 493 (51.14%) with AD dementia. Among the responses, 122 (12.66%) were completed by patients themselves, while 842 (87.34%) were filled out by their caregivers. Most patients, 81.54% (786) were living with family members, 58 (6.02%) resided in professional eldercare facilities, and 34 (3.53%) were in specialized medical institutions. Geographically, 53.7% (518) of the patients were from eastern China, 10.3% (99) from central China, 5.5% (53) from northeastern China, and 30.5% (294) from western China. Additionally, 44.7% (431) of the respondents held a bachelor's degree or higher.

Among the diagnostic tests performed, 136 patients (14.11%) underwent APOE gene polymorphism testing. Of these, 25 were identified as ε4/ε4 carriers; 43 carried ε2/ε4 or ε3/ε4 genotypes; and 68 had ε2/ε2, ε2/ε3, or ε3/ε3 genotypes. A total of 228 (23.65%) patients were diagnosed with AD according to the amyloid/tau/neurodegeneration (ATN) framework, of which 85 patients were diagnosed via cerebrospinal fluid (CSF) analysis, and 140 patients were diagnosed via amyloid positron emission tomography (PET) scan. For intervention and treatment, 155 (16.08%) patients received lifestyle interventions; 409 (42.47%) patients were treated with donepezil hydrochloride tablets, 419 (43.47%) with memantine hydrochloride tablets, 52 (5.39%) with exelon (rivastigmine tartrate capsules), 32 (3.32%) with galantamine hydrobromide tablets, 20 (2.07%) with huperzine A tablets, 223 (23.13%) with mannitol sodium capsules, and 35 (3.63%) received amyloid beta (Aβ) monoclonal antibody (lecanemab) treatment.

3.2. Food consumption and nutrient intake among patients with cognitive impairment in China

This survey found that the median daily intake per person among patients with cognitive impairment in China was 132.26 ± 71.23 g of cereals and tubers, 159.68 ± 38.98 g of vegetables, 20.44 ± 27.98 g of fish and shrimp, 67.04 ± 73.06 g of fruit, 68.54 ± 59.64 g of meat and poultry, 32.67 ± 25.08 g of eggs, 18.40 ± 18.39 g of soybeans, and 103.74 ± 115.97 g of dairy products per person per day (Table 1).

TABLE 1.

The average daily food intake among patients with cognitive impairment.

Diet component	The average daily intake per person	Recommended intake a	CNHS (2015–2017) b
Cereals and tubers	132.26 ± 71.23	250–400	316.8
Whole grains	12.28 ± 18.36	50–150	17.3
Tubers	20.41 ± 33.07	50–100	41.3
Vegetables	159.68 ± 38.98	300–500	255.9
Fruits	67.04 ± 73.06	200–350	30.9
Meat and poultry	68.54 ± 59.64	80–150	66.5
Fish and shrimp	20.44 ± 27.98	40–75	22.1
Eggs	32.67 ± 25.08	40–75	19.1
Soybean and soy products	18.40 ± 18.39	25–35	9.9
Milk and milk products	103.74 ± 115.97	300–500	23.2
Nuts	12.86 ± 21.29	10	3.5
Snacks or desserts	16.32 ± 26.49	–	7.3
Alcohol	14.14 ± 69.30	Female<15 g Male<25 g	2.8

Note: Values for the study population are mean ± SD; recommended intake and CNHS data are reference values.

^a Recommendations in the Chinese Dietary Guidelines for Older Adults (2022).

^b Dietary data of adults aged ≥ 60 from the CNHS (2015–2017).

Abbreviations: CNHS, China Nutrition and Health Surveillance; SD, standard deviation.

The average daily intake of several food groups among patients with cognitive impairment was below the recommended levels outlined in the Chinese Dietary Guidelines for Older Adults (2022), particularly for cereals and tubers, vegetables, fruits, meat and poultry, fish and shrimp, eggs, soybean products, and dairy. Additionally, the average intake of cereals and tubers, whole grains, and vegetables in the present study population was lower than the averages reported in the CNHS (2015–2017). Conversely, higher average intakes of fruits, dairy products, alcohol, and snacks were observed relative to national surveillance data. These findings reflect descriptive comparisons without statistical inference and are presented in Table 1 and Figure 1 for visual interpretation.

FIGURE 1.

Radar chart comparing the average daily food intake of patients with cognitive impairment to recommended levels in the Chinese Dietary Guidelines for Older Adults (2022) and dietary data from the China Nutrition and Health Surveillance (2015–2017). For visual comparison, all intake values were expressed as percentages relative to the minimum recommended intake values from the Chinese Dietary Guidelines for Older Adults (2022)

The average daily energy intake per person among patients with cognitive impairment in China was 1131.59 ± 486.30 kcal, which is significantly below the dietary recommended intake for older adults in China and lower than CNHS (2015–2017). The average daily intake of protein, fat, and carbohydrates per person was 50.96 g, 37.23 g, and 151.46 g, respectively. The contribution of protein, fat, and carbohydrates to total energy was 18.01%, 29.61%, and 53.54%, respectively, which generally aligns with recommended energy distribution standards (Table 2). Additionally, the proportion of energy derived from saturated fat reached 22.7%, which is notably higher than the recommended value. The intake of dietary fiber, zinc, vitamin B1, vitamin B2, folate, and soy isoflavones was considerably below the recommended levels, with only iron and vitamin E meeting the recommended standards.

TABLE 2.

Energy and major nutrient intake among patients with cognitive impairment in China.

Nutrient	RNIs/AIs a	The average daily intake per person	CNHS (2015–2017) b
Energy (kcal)	Male: 1900 female: 1550	1131.59 ± 486.30	1774.4
Protein (g)	Female: 62 g; male: 72 g	50.96 ± 24.29	52.9
Fat (g)	20%–30%	37.23 ± 20.78	67.2
Carbohydrate (g)	120	151.46 ± 69.17	241.2
Fiber (g)	25–30	7.62 ± 4.67	9.6
Saturated fat (g)	<10%	28.56 ± 18.13	–
Iron (mg)	10	15.82 ± 8.09	18.7
Zinc (mg)	8.5	6.49 ± 3.07	9.0
Selenium (µg)	50	36.22 ± 19.67	35.8
Vitamin B1 (mg)	1.0	0.64 ± 0.33	0.7
Vitamin B2 (mg)	1.0	0.59 ± 0.29	0.6
Folic acid (µg)	320	207.19 ± 139.43	–
Vitamin E (mg)	14	14.29 ± 9.29	32.4
Vitamin A (µgRAE)	640	376.42 ± 272.57	396.6

Note: Values for the study population are mean ± SD; RNIs/AIs and CNHS data are reference values.

Abbreviations: AI, adequate intake; CNHS, China Nutrition and Health Surveillance; RNI, recommended nutrient intake.

^a Reference values were obtained from the Dietary Reference Intakes for Chinese Residents (2023), published by the Chinese Nutrition Society.

^b Dietary data of adults aged ≥ 60 from the CNHS (2015–2017).

3.3. Characteristics of patients with cognitive impairment by DII quartiles

The average DII value for patients with cognitive impairment was 2.39 ± 1.71, with a range from –3.60 to 5.71. There were significant differences in disease duration across the DII quartiles (P = 0.044, Kruskal–Wallis test), and in the distribution of cognitive impairment stages (P < 0.001, chi‐squared test). Additionally, the intake of energy, protein, carbohydrates, and other nutrients in the pro‐inflammatory diet (Q4) was significantly lower than that in the anti‐inflammatory diet group (Q1; P < 0.001, ANOVA with Bonferroni post hoc test), with energy and macronutrient intake being only 50% of that of the anti‐inflammatory diet  group (Tables 3 and 4).

TABLE 3.

Sociodemographic and clinical characteristics of participants by DII quartiles.

Variable	Q1	Q2	Q3	Q4	P
Years of cognitive impairment diagnosis, years	2.92 ± 3.57	2.42 ± 3.16	2.74 ± 3.06	3.26 ± 3.36	0.044
Living pattern, n (%)					0.239
With family members	206 (85.48)	197 (81.74)	189 (78.42)	194 (80.50)
Alone or at nursing home	35 (14.52)	44 (18.26)	52 (21.58)	47 (19.50)
Stage of cognitive impairment, n (%)					<0.001
SCD	82 (34.03)	90 (37.34)	86 (35.69)	63 (26.14)
MCI	41 (17.01)	48 (19.92)	37 (15.35)	24 (9.96)
Mild dementia	32 (13.28)	34 (14.11)	29 (12.03)	24 (9.96)
Moderate dementia	51 (21.16)	40 (16.60)	53 (21.99)	85 (35.27)
Severe dementia	35 (14.52)	29 (12.03)	36 (14.94)	45 (18.67)
Lifestyle or medication treatment, n (%)					0.153
Yes	77 (31.95)	86 (35.68)	78 (32.37)	63 (26.14)
No	164 (68.05)	155 (64.32)	163 (67.63)	178 (73.86)
Region, n (%)					0.129
Northeast	9 (3.73)	21 (8.71)	10 (4.15)	13 (5.39)
Eastern	126 (52.28)	133 (55.19)	135 (56.02)	124 (51.45)
Central	26 (10.79)	15 (6.22)	28 (11.62)	30 (12.45)
Western	80 (33.20)	72 (29.88)	68 (28.22)	74 (30.71)
APOE haplotype, n (%)					0.026
Untested	211 (87.55)	218 (90.46)	206 (85.48)	193 (80.08)
1 (ε4/ε4)	8 (3.32)	2 (0.83)	9 (3.73)	6 (2.49)
2 (ε2/ε4, ε3/ε4)	10 (4.15)	10 (4.15)	10 (4.15)	13 (5.39)
3 (ε2/ε2, ε2/ε3 or ε3/ε3)	12 (4.98)	11 (4.56)	16 (6.64)	29 (12.03)

Note: Table values are mean  ± SD or N (%). P values based on Kruskal–Wallis test or chi‐squared test.

Abbreviations: APOE, apolipoprotein E; DII, Dietary Inflammatory Index; MCI, mild cognitive impairment; SCD, subjective cognitive decline; SD, standard deviation.

TABLE 4.

Comparison of nutrient intake across DII quartiles.

Nutrient	Q1	Q2	Q3	Q4	P
DII	0.16 ± 1.01	1.87 ± 0.34	2.97 ± 0.33	4.54 ± 0.51	<0.001
Energy (kcal)	1565.15 ± 432.84	1200.21 ± 370.51	1016.82 ± 363.52	843.65 ± 354.30	<0.001
Protein (g)	75.99 ± 22.85	54.80 ± 15.52	44.75 ± 16.31	32.79 ± 13.30	<0.001
Fat (g)	56.33 ± 21.84	39.97 ± 15.11	32.42 ± 14.70	23.52 ± 12.46	<0.001
Carbohydrate (g)	196.50 ± 67.77	157.77 ± 59.61	138.57 ± 57.54	126.20 ± 59.65	<0.001
Fiber (g)	13.05 ± 4.72	7.67 ± 2.76	6.00 ± 2.75	4.44 ± 2.26	<0.001
Saturated fat (g)	36.36 ± 18.79	28.24 ± 17.46	27.69 ± 17.09	24.34 ± 16.07	<0.001
Iron (mg)	24.64 ± 7.68	16.71 ± 4.84	13.34 ± 4.98	9.96 ± 4.71	<0.001
Zinc (mg)	9.99 ± 2.69	6.86 ± 1.75	5.54 ± 1.86	4.14 ± 1.69	<0.001
Selenium (µg)	52.25 ± 21.29	38.72 ± 14.71	32.90 ± 14.72	24.28 ± 13.29	<0.001
Vitamin B1 (mg)	0.93 ± 0.29	0.68 ± 0.25	0.57 ± 0.29	0.43 ± 0.21	<0.001
Vitamin B2 (mg)	0.90 ± 0.29	0.60 ± 0.18	0.50 ± 0.21	0.41 ± 0.19	<0.001
Folic acid (µg)	376.68 ± 150.73	214.06 ± 67.75	153.94 ± 64.23	102.25 ± 47.86	<0.001
Vitamin E (mg)	24.71 ± 9.77	16.27 ± 5.41	11.48 ± 4.09	5.99 ± 2.45	<0.001
Vitamin A (µgRAE)	682.03 ± 307.70	355.02 ± 168.94	282.35 ± 60.74	218.66 ± 131.47	<0.001
PUFA (mg)	11.20 ± 4.64	7.77 ± 2.82	5.88 ± 2.38	3.65 ± 1.86	<0.001
MUFA (mg)	25.26 ± 11.96	17.46 ± 7.93	13.91 ± 7.23	9.66 ± 6.57	<0.001

Note: Table values are mean ± SD. P values based on Kruskal–Wallis test or analysis of variance.

Abbreviations: DII, Dietary Inflammatory Index; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; SD, standard deviation.

3.4. Comparison of dietary intake across different stages of cognitive impairment

Table 5 summarizes the daily nutrient intake of participants stratified by cognitive impairment stage. DII values differed significantly across cognitive impairment stages (P < 0.001, Kruskal–Wallis test), with higher values observed from SCD to severe dementia, indicating greater dietary inflammatory potential with increasing cognitive impairment severity. Similarly, significant differences in carbohydrate (P = 0.005, ANOVA), saturated fat (P < 0.001, ANOVA), vitamin E (P = 0.033, ANOVA), vitamin D (P = 0.002, ANOVA), and isoflavones (Pp = 0.014, ANOVA) intake were observed across cognitive impairment stages. No significant differences were detected in total energy intake, protein, fiber, or most micronutrients including iron, zinc, selenium, folic acid, and vitamin A (P > 0.05, ANOVA).

TABLE 5.

Nutrient intake across stages of cognitive impairment.

Nutrient	SCD	MCI	Mild dementia	Moderate dementia	Severe dementia	P
DII	2.24 ± 1.64	2.11 ± 1.54	2.21 ± 1.70	2.78 ± 1.77	2.50 ± 1.85	<0.001
Energy (kcal)	1111.59 ± 482.28	1197.82 ± 491.08	1184.09 ± 447.17	1170.65 ± 442.34	1167.92 ± 449.71	0.300
Protein (g)	51.40 ± 24.28	55.41 ± 25.36	53.33 ± 22.50	50.33 ± 20.98	51.89 ± 24.21	0.297
Fat (g)	37.39 ± 20.79	40.86 ± 21.94	38.67 ± 18.43	37.03 ± 19.41	37.76 ± 20.53	0.417
Carbohydrate (g)	143.70 ± 67.33	154.85 ± 66.94	158.93 ± 67.35	164.00 ± 67.35	161.11 ± 63.79	0.005
Fiber (g)	7.41 ± 4.49	8.06 ± 4.49	8.04 ± 4.55	7.80 ± 4.63	8.13 ± 4.94	0.434
Saturated fat (g)	25.64 ± 18.32	30.88 ± 17.88	28.95 ± 16.86	31.56 ± 17.26	31.57 ± 17.80	<0.001
Iron (mg)	15.55 ± 7.99	17.36 ± 8.11	16.47 ± 7.49	15.87 ± 7.37	16.51 ± 8.33	0.181
Zinc (mg)	6.45 ± 3.03	6.98 ± 3.05	6.98 ± 3.01	6.45 ± 2.77	6.70 ± 3.03	0.213
Selenium (µg)	36.24 ± 20.56	38.13 ± 19.10	39.12 ± 19.08	36.35 ± 17.81	37.06 ± 18.46	0.604
Vitamin B1 (mg)	0.64 ± 0.32	0.70 ± 0.34	0.68 ± 0.35	0.64 ± 0.29	0.64 ± 0.30	0.282
Vitamin B2 (mg)	0.58 ± 0.30	0.63 ± 0.28	0.63 ± 0.27	0.60 ± 0.27	0.63 ± 0.30	0.164
Folic acid (µg)	212.48 ± 142.86	220.09 ± 32.00	226.26 ± 139.45	191.99 ± 124.25	220.70 ± 150.65	0.127
Vitamin E (mg)	14.51 ± 8.60	16.62 ± 10.18	14.64 ± 9.42	13.54 ± 8.73	14.46 ± 9.51	0.033
Vitamin A (µgRAE)	366.46 ± 279.43	397.58 ± 245.97	415.13 ± 287.52	372.32 ± 246.79	405.12 ± 296.16	0.331
PUFA (mg)	7.07 ± 4.03	7.92 ± 4.75	7.22 ± 3.76	6.69 ± 3.98	7.02 ± 4.31	0.081
MUFA (mg)	16.62 ± 10.75	17.70 ± 10.94	17.09 ± 9.77	15.68 ± 10.03	16.29 ± 10.06	0.428
Vitamin D (µg)	0.62 ± 0.51	0.75 ± 0.50	0.70 ± 0.51	0.78 ± 0.50	0.78 ± 0.50	0.002
Isoflavones (mg)	3.61 ± 3.18	4.06 ± 3.85	3.38 ± 3.51	2.89 ± 3.09	3.28 ± 3.34	0.014

Note: Table values are mean ± SD. P values based on Kruskal–Wallis test or analysis of variance.

Abbreviations: DII, Dietary Inflammatory Index; MCI, mild cognitive impairment; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; SCD, subjective cognitive decline; SD, standard deviation.

3.5. Association of DII and dietary patterns with cognitive impairment severity

Supplementary Table S1 in supporting information shows the four dietary patterns identified by PCA and the factor loadings for each food group. These four dietary patterns ‐ the Eastern balanced dietary pattern, high‐quality protein dietary pattern, high‐snack and high‐alcohol dietary pattern, and high cereal and snack dietary pattern‐explained 20.42%, 12.11%, 11.00%, and 9.39% of the variance in the response variable, respectively. The first dietary pattern, called the “Eastern balanced dietary pattern,” is characterized by high consumption of vegetables, fruit, fish and shrimp, soybeans, nuts, and cereals. The second dietary pattern, the “high‐quality protein dietary pattern,” was characterized by high consumption of meat and poultry, soybeans, fish, and shrimp. the third pattern is called the “high‐snack and high‐alcohol dietary pattern” and is characterized by high consumption of snacks or desserts and alcohol. The fourth dietary pattern, called the “high cereal and snack dietary pattern,” is characterized by high consumption of refined grains and snacks or desserts.

As shown in Figure 2, DII was negatively associated with the “Eastern balanced dietary pattern,” “high‐quality protein dietary pattern,” and “high cereal and snack dietary pattern,” but positively associated with “high‐snack and high‐alcohol dietary pattern.” A positive correlation was also observed between DII and the stage of cognitive impairment.

FIGURE 2.

Correlates of DII, dietary pattern and cognitive impairment. Red color indicates inverse associations, blue indicates positive associations. Pearson correlation coefficient, two sided 95%, ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001. DII, Dietary Inflammatory Index

In the unadjusted model (Model 1), higher DII was significantly associated with more severe cognitive impairment (β = 0.179, 95% confidence interval [CI]: 0.077–0.280, P trend < 0.001, ordinal logistic regression). This association remained significant after adjusting for APOE genotype and years of cognitive impairment diagnosis in Model 2 (β = 0.130, 95% CI: 0.022–0.238, P trend = 0.018, ordinal logistic regression), indicating an independent association between dietary inflammatory potential and the severity of cognitive impairment (Table 6).

TABLE 6.

Ordinal logistic regression models of DII and dietary patterns associated with cognitive impairment severity.

Dietary patterns	Model 1			Model 2
	β	(95% CI)	P trend	β	(95% CI)	P trend
DII	0.179	0.077, 0.280	<0.001	0.130	0.022, 0.238	0.018
Eastern balanced dietary pattern	0.043	−0.058, 0.144	0.401	0.022	−0.086, 0.129	0.693
High‐quality protein dietary pattern	−0.365	−0.469, −0.262	<0.001	−0.269	−0.378, −0.159	<0.001
High‐snack and high‐alcohol dietary pattern	0.035	−0.065, 0.136	0.491	−0.208	−0.136, 0.079	0.604
High cereal and snack dietary pattern	0.200	0.099, 0.302	<0.001	0.196	0.088, 0.304	<0.001

Note: Model 1 includes each dietary variable separately. Model 2 adjusts for APOE genotype and years of cognitive impairment diagnosis. P trend derived from modeling ordinal variables as continuous predictors.

Abbreviations: APOE, apolipoprotein E; CI, confidence interval; DII, Dietary Inflammatory Index.

Among dietary patterns, the high‐quality protein dietary pattern was inversely associated with cognitive impairment severity in both models (Model 1: β = −0.365, 95% CI: −0.469 to −0.262; Model 2: β = −0.269, 95% CI: −0.378 to −0.159; both P trend < 0.001, ordinal logistic regression). In contrast, the high cereal and snack dietary pattern showed a positive association with cognitive impairment (Model 1: β = 0.200, 95% CI: 0.099–0.302; Model 2: β = 0.196, 95% CI: 0.088–0.304; both P trend < 0.001, ordinal logistic regression). The Eastern balanced dietary pattern and the high‐snack and high‐alcohol dietary pattern were not significantly associated with the severity of cognitive impairment in either model (Table 6).

4. DISCUSSION

This study is the first to systematically investigate the dietary and nutritional intake of patients with cognitive impairment in China. Dietary data were interpreted using two complementary reference frameworks: the Chinese Dietary Guidelines for Older Adults, which outline nutrient intake recommendations, and findings from the CNHS, which provide nationally representative data on actual dietary practices among older adults. The integration of recommended standards with surveillance data allowed for the assessment of both dietary adequacy and deviations from typical intake patterns. This approach provides a clearer understanding of dietary characteristics linked to cognitive impairment. Most patients with cognitive impairment in China showed insufficient intake of key nutrients, including energy, carbohydrates, vitamin A, vitamin B1, vitamin B2, and folate. Mean energy intake was substantially lower than national guidelines and the average intake reported for older adults in 2015, ²⁶ with 78.53% of patients consuming < 1500 kcal/day.

In this study, higher DII scores reflected an overall deficiency in nutrient intake, rather than excessive consumption of “pro‐inflammatory” foods. Insufficient energy intake is commonly observed among individuals with cognitive impairment in China, which exacerbates the risk of energy‐protein malnutrition. Similarly, nutrition risk and malnutrition are prevalent among individuals with cognitive impairment worldwide. Evidence suggests that impaired appetite, dysphagia, and disrupted eating behaviors contribute to weight loss and malnutrition, particularly in individuals with AD. ⁶ , ²⁸ Cognitive decline often leads to a decreased interest in food, which further exacerbates inadequate nutrient intake. ²⁹ Additionally, swallowing difficulties, a prevalent symptom in dementia, can limit food consumption and contribute to nutritional deficiencies. ⁶ , ³⁰ Furthermore, while protein intake was found to be insufficient for a significant proportion of patients, with only 45.4% consuming > 50 g per day, 43.88% of patients had a total fatenergy ratio within the recommended intake range. However, only 14.94% had a saturated fat–energy ratio within the recommended range, indicating that the majority of patients consumed an unhealthy level of saturated fat. Dietary fat is essential as an energy source, a component of cell membranes, and a precursor for signaling molecules. A higher proportion of energy derived from saturated fats has been shown to increase adiposity, promote macrophage infiltration, elevate inflammation, and contribute to insulin resistance, all of which can impair the lipid profile. ³¹

Previous studies suggested that pro‐inflammatory diets exacerbate cognitive decline. ⁹ , ³² , ³³ , ³⁴ Furthermore, a trend was observed linking the DII to a reduction in hippocampal gray matter volume. ¹¹ , ¹² However, few studies have explored whether this pro‐inflammatory diet is due to excessive intake of pro‐inflammatory foods or insufficient intake of anti‐inflammatory foods. A study using the National Health and Nutrition Examination Survey (NHANES) dataset on elderly individuals in the United States found that the pro‐inflammatory diet was primarily due to excessive consumption of pro‐inflammatory foods, including high levels of energy, fat, and carbohydrates, while intake of anti‐inflammatory foods, such as protein and dietary fiber, was relatively low. ⁹ In contrast, we found that the pro‐inflammatory diet in patients with cognitive impairment in China was not due to the consumption of excessively high levels of “pro‐inflammatory” foods, but rather resulted from an overall insufficient intake of nutrients. Specifically, the intake of energy, protein, carbohydrates, and other nutrients in the pro‐inflammatory diet group was approximately half of that in the anti‐inflammatory diet group. Meanwhile, a positive correlation was observed between the DII and the severity of the disease, suggesting that as the disease progresses, nutrient intake among patients with cognitive impairment tends to decline, thereby intensifying the inflammatory potential of their diet. As cognitive impairment progresses, particularly in AD, individuals may experience lapses in food intake or forget to eat meals, make inappropriate food choices, and have difficulty recognizing certain types of foods. ⁶ , ²⁸ , ³⁵ , ³⁶ This can create a vicious cycle, in which insufficient nutrient intake contributes to a pro‐inflammatory diet, which exacerbates systemic inflammation, promoting the onset of malnutrition ³⁷ and further negatively affecting disease progression and diminishing quality of life.

Further analysis identified four distinct dietary patterns among patients with cognitive impairment. The findings revealed a strong negative correlation between the Eastern balanced dietary pattern and the DII, while the high‐snack and high‐alcohol dietary pattern showed a positive association. The Eastern balanced dietary pattern, characterized by high consumption of vegetables, fruits, fish and shrimp, soybeans, nuts, and cereals, is consistent with dietary patterns previously associated with improved cognitive outcomes in older adults. A 15‐year follow‐up study from the Chinese Longitudinal Healthy Longevity Survey revealed that regular consumption of fruits, vegetables, meat, and soybean‐based products may be beneficial in maintaining cognitive health in aging populations. ³⁸ Conversely, the high‐snack and high‐alcohol dietary pattern, which includes high intake of processed snacks and alcoholic beverages, was positively correlated with DII, indicating a pro‐inflammatory profile. High consumption of snacks, particularly those high in added sugars and unhealthy fats, can elevate inflammation, which may contribute to cognitive decline. ³⁸ , ³⁹ A cohort study of older adults in China also found a potential causal association between alcohol consumption and cognitive impairment. ⁴⁰

Beyond its correlation with DII, we conducted ordinal logistic regression to examine associations between dietary patterns and the severity of cognitive impairment. The high‐quality protein dietary pattern was associated with lower cognitive impairment severity, while the high cereal and snack dietary pattern was linked to increased severity, independent of APOE genotype and disease duration. These findings suggest that dietary patterns rich in high‐quality protein (such as eggs, lean meat, dairy, and soy) may offer protective effects, whereas diets characterized by excessive intake of low‐quality carbohydrates may contribute to poorer cognitive outcomes.

Recent epidemiological evidence supports the protective role of high‐quality protein intake in cognitive function. A large prospective cohort study involving > 75,000 US adults demonstrated that higher long‐term protein intake, especially from plant‐based and lean animal sources (e.g., fish, bean/legumes, poultry), was significantly associated with a reduced risk of subjective cognitive decline.⁴¹ The protective role of dietary protein may involve multiple mechanisms. Adequate protein intake supports neurotransmitter synthesis, synaptic function, and muscle maintenance—all of which are closely associated with cognitive performance and aging‐related brain health. ⁴² , ⁴³ Moreover, high‐quality protein sources are rich in micronutrients such as vitamin B12, iron, and n‐3 polyunsaturated fatty acids, which have demonstrated neuroprotective effects. ⁴²

Conversely, excessive intake of low‐quality carbohydrates—such as added sugars and ultra‐processed snacks—may accelerate cognitive deterioration through metabolic and inflammatory pathways. A Chinese cohort study found that high intake of low‐quality carbohydrates was linked to accelerated cognitive decline, while isocaloric substitution with animal protein mitigated the risk. ⁴⁴ Diets with a high glycemic index may induce postprandial hyperglycemia and oxidative stress, which are implicated in neurodegeneration and cognitive decline. ⁴⁴ , ⁴⁵ Our findings emphasize the importance of dietary patterns that prioritize high‐quality protein and reduce low‐quality carbohydrate intake to support cognitive health in older adults. Additionally, the identification of these patterns underscores the potential for tailored nutritional interventions in patients with cognitive impairment. Promoting diets with adequate energy, high‐quality protein, and antioxidants—while limiting processed snacks—may reduce dietary inflammation and support cognitive health. Educating patients and their caregivers about the components and benefits of an anti‐inflammatory diet could serve as an effective strategy to optimize nutritional intake and potentially mitigate cognitive decline.

This study has several limitations. First, as the primary aim of the study was to assess dietary intake, and given the complexity of the food frequency questionnaire, efforts were made to reduce respondent burden and avoid confusion—particularly considering that the majority of questionnaires were completed by caregivers on behalf of patients. As a result, only basic information about the respondent was collected, and patient‐level demographic variables such as age and sex were omitted. This limited adjustment for these potential confounders, which may influence the interpretation of the independent association between dietary inflammatory potential and cognitive impairment severity. Second, the cross‐sectional design of this study restricts our ability to establish causal relationships between dietary patterns and cognitive impairment. Longitudinal research is necessary to gain deeper insights into how dietary factors influence cognitive health as the disease advances. Third, dietary assessments relied on self‐reported data, which may be prone to recall bias and inaccuracies. Participants might have underreported or overreported their food intake, potentially resulting in misclassification of dietary patterns. Fourth, although we identified distinct dietary patterns, the study did not consider other lifestyle factors, such as physical activity and social engagement, that may also affect cognitive health. Future research should examine these confounding variables for a more comprehensive understanding of the relationship between dietary intake and cognitive impairment.

In conclusion, inadequate intake of essential nutrients—particularly energy and high‐quality protein—remains a significant issue among individuals with cognitive impairment, contributing to increased dietary inflammatory potential. Our findings suggest that adherence to a high‐quality protein dietary pattern is associated with lower severity of cognitive impairment, while the high cereal and snack dietary pattern is associated with greater severity. These results highlight the need for dietary strategies that ensure sufficient protein intake while limiting low‐quality carbohydrates. This study underscores the importance of targeted nutritional interventions and the integration of specific dietary guidelines into treatment protocols for cognitive impairment, ultimately improving the quality of life and health outcomes for individuals with cognitive decline.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest. Author disclosures are available in the supporting information.

CONSENT STATEMENT

Informed consent was obtained from all participants or their caregivers.

Supporting information

Supporting Information

Supporting Information

Bian D, Liang F, You Y, et al. Evaluating dietary quality and dietary inflammatory potential in cognitive impairment patients in China. Alzheimer's Dement. 2025;21:e70345. 10.1002/alz.70345

DATA AVAILABILITY STATEMENT

Data are available upon reasonable request.