Associations of starchy and non-starchy vegetable intake with depression, anxiety, stress, chronotype, and sleep quality among iranian women with overweight or obesity: a cross-sectional study

Abstract

Background

Starchy vegetables may influence psychological well-being, circadian alignment, and sleep quality through their glycemic characteristics and related metabolic effects. In contrast, non‑starchy vegetables, which differ substantially in nutrient density and bioactive compounds, may affect these outcomes through alternative physiological pathways. Despite these distinct nutritional profiles, evidence comparing their separate associations with mental health and sleep-related patterns is limited. This study explored these relationships among overweight and obese women.

Methods

In this cross-sectional study, 391 overweight and obese women aged 18–48 years were recruited using a cluster sampling strategy from community health centres. Data collection was conducted through structured face‑to‑face interviews performed by trained nutritionists. Mental health indicators (depression, anxiety, and stress) were assessed using the Persian DASS‑21, sleep quality was evaluated using the PSQI, and circadian preference was measured using the MEQ. Dietary intake was obtained through a validated 147‑item food frequency questionnaire (FFQ). Anthropometric measurements, blood samples, biochemical markers, physical activity, and demographic information were collected following standardized procedures.

Results

The study included 391 women, with a mean age of 38.08 ± 8.61 years in the high non-starchy vegetable intake group and 35.31 ± 9.59 years in the low non-starchy vegetable intake group, all classified as overweight or obese. Participants with higher non‑starchy vegetable intake exhibited significantly lower odds of depression (adjusted OR = 0.40, 95% CI = 0.20–0.79, P < 0.001) and stress (adjusted OR = 0.30, 95% CI = 0.14–0.85, P = 0.02). Higher non‑starchy vegetable consumption was also associated with improved sleep quality (adjusted OR = 0.38, 95% CI = 0.19–0.77, P = 0.007). Furthermore, increased starchy vegetable intake was inversely associated with the morning-type chronotype (adjusted OR = 0.27, 95% CI = 0.09–0.83, P = 0.02) after full adjustment.

Conclusions

Higher non-starchy vegetable intake was associated with lower odds of depression, stress, and poor sleep quality, while greater starchy vegetable consumption was linked to a lower likelihood of morning-type chronotype. Increasing non-starchy vegetable intake may support psychological well-being and sleep health, though longitudinal studies are needed to confirm causality.

Introduction

Obesity and overweight are recognized as major global health burdens due to their strong association with a wide range of chronic diseases [1]. These conditions arise from a complex interplay of genetic susceptibility, environmental exposures, and underlying medical problems [2]. According to the World Health Organization (WHO), by 2025, nearly 167 million individuals will experience health impairments linked to excess body weight [3]. In Iran, obesity affects a considerable portion of the population, with prevalence estimates ranging from 15.8% to 36.5% among men and from 31.3% to 76.7% among women [4]. Obesity has been associated with mental disorders such as depression and anxiety [5] and disruptions in circadian functioning [5, 6].

Multiple pathways have been proposed to explain the link between obesity and depressive symptoms. One study reported a higher prevalence of depressive disorders among young adults with obesity and suggested that hypothalamic inflammation—particularly in the left hypothalamus—may contribute to the co-occurrence of obesity and depression [7]. Other evidence shows that obesity and depression can jointly impair sleep quality, indicating a potential interrelationship between these conditions [8]. Psychological consequences of obesity, including reduced self-esteem, may further contribute to depression [9]. Research on centenarians also demonstrated that various obesity phenotypes—such as visceral adiposity measured through waist circumference (WC) and waist-to-calf circumference ratio, as well as peripheral obesity assessed via calf circumference—exert different influences on depressive symptoms [10]. Chronic low-grade systemic inflammation associated with obesity, characterised by elevated pro-inflammatory cytokines, has been implicated in the pathophysiology of depression, providing a potential biological link between these conditions [11]. Moreover, obesity and metabolic syndrome can lead to disturbances in daily behavioral patterns, creating misalignment in circadian rhythms and promoting the development or progression of sleep disorders [12].

The circadian rhythm represents an individual’s internal timing system governing daily patterns related to sleep, eating, and activity within a 24-hour cycle [13]. Recent evidence has shown that circadian preferences are closely linked to dietary behaviors, meal timing, eating patterns, and obesity risk [14]. For instance, adults with an evening chronotype tend to have higher body mass index (BMI) and WC than those with a morning chronotype [15]. The morning chronotype is characterized by a preference for earlier sleep and wake times, with peak alertness occurring in the early part of the day, whereas the evening chronotype involves a preference for later sleep and wake times, with peak alertness in the late afternoon or evening. Morning chronotypes typically exhibit more regular sleep-wake patterns and have been associated with better metabolic health, lower obesity risk, and improved psychological well-being compared with evening chronotypes, who are more prone to circadian misalignment, unhealthy eating behaviors, and higher rates of depression and anxiety. Sleep disturbances, meanwhile, have become increasingly widespread worldwide [16, 17]. Adequate sleep plays a crucial role in metabolic and hormonal regulation [18, 19], and both short and long sleep durations, as well as poor sleep quality, have been associated with increased risks of overall and abdominal obesity [20, 21]. However, the strength and direction of the relationship between obesity, sleep duration, and sleep quality may vary across contexts [22, 23].

Dietary intake, particularly vegetable consumption, has been highlighted as a potentially modifiable factor influencing mental health and sleep [24]. Previous studies have shown that vegetable intake may enhance psychological well-being [25] and sleep duration [26]. A longitudinal investigation in Thailand among pre-diabetic adults reported that frequent vegetable consumption improved mental health indicators linked to quality of life [27]. Similarly, research in Spain involving overweight and obese adults found that higher vegetable intake was associated with greater life satisfaction [28]. Moreover, adherence to plant-rich dietary patterns, such as the Mediterranean diet, has been linked to improved sleep quality and duration, particularly in older adults. These findings are supported by a recent systematic review and meta-analysis reporting that higher adherence to the Mediterranean diet was associated with a 14% lower risk of insomnia [29]. likely due to complex carbohydrates that influence tryptophan metabolism, and bioactive compounds, including polyphenols and fibre, which support circadian and inflammatory pathways [30]. Despite these findings, the effects of different types of vegetables on mental health and sleep remain poorly understood, and to our knowledge, limited evidence has directly compared the influence of starchy and non-starchy vegetables on depression, anxiety, circadian rhythm, or sleep quality.

Vegetables are generally classified as starchy or non-starchy according to their carbohydrate and energy content. Starchy vegetables, such as potatoes, corn, peas, and squash, are richer in calories and carbohydrates, whereas non-starchy vegetables like leafy greens, broccoli, cauliflower, peppers, and tomatoes are lower in calories but higher in fiber, vitamins, and minerals. Evidence suggests that non-starchy vegetables may confer greater health benefits, including reduced mortality risk [31], and replacing starchy with non-starchy vegetables may enhance adherence to a Mediterranean dietary pattern, which is linked to lower chronic disease risk [32]. While some research indicates that starchy vegetables may alleviate depressive symptoms in university students [33], green leafy vegetables may improve mental health by increasing serotonin levels, a key regulator of mood and sleep [25, 34].

Despite these findings, limited evidence has directly compared starchy and non-starchy vegetables regarding their associations with mental health, circadian rhythm, and sleep quality. Given the higher prevalence of depression and sleep disturbances among women, along with sex-specific hormonal and metabolic influences on circadian and mood regulation [11, 35]. Therefore, the present study aimed to investigate which type of vegetables is more strongly associated with mental well-being and sleep quality in overweight and obese Iranian women.

Methods

Study population

This cross-sectional study included 391 overweight or obese women. Inclusion criteria were: aged 18–48 years with a BMI of 25–40 kg/m², recruited from 20 health centers in Tehran using a multi-stage cluster random sampling method. The health centers were randomly selected from five geographically distinct districts of Tehran (north, south, east, west, and center), with four health centers chosen from each district. From each health center, approximately 19–20 eligible women were randomly selected to participate. The cluster sampling design was accounted for in the statistical analyses by including health center as a random effect in the regression models. Written informed consent was obtained from all participants before enrollment. Exclusion criteria were: drug or alcohol consumption; energy intake < 800 or > 4200 kcal/day; history of hypertension, cardiovascular disease, type 2 diabetes, hepatic or renal failure; use of medications including oral contraceptives or antidepressants in the past year; smoking; pregnancy or lactation; menopause; adherence to specific diets; or weight fluctuation (≥ 5% change in body weight) during the past six months. Individuals with a history of bariatric surgery, weight‑loss surgical procedures, or the use of pharmacological or non‑oral agents known to affect body weight (such as injectable weight‑loss medications) were also excluded. The study protocol was approved by the Human Ethics Committee of Tehran University of Medical Sciences (Ethics No. IR.TUMS.VCR.REC.1398.692).

Demographic and physical activity assessment

Demographic data (age, education, occupation, marital status, and economic status) were collected via questionnaire. Physical activity was assessed using the International Physical Activity Questionnaire (IPAQ) and expressed as MET-hours/week [36].

Assessment of depression, anxiety, and stress

Depression, anxiety, and stress were assessed using the validated Persian version of the 21-item Depression, Anxiety, and Stress Scale (DASS-21) [37]. The questionnaire includes 21 items in total, with seven items for each subscale: depression, anxiety, and stress. Each item is scored from 0 (never) to 3 (always). For depression, scores are classified as normal (0–9), mild [10–13], moderate [14–20], severe [21–27], and very severe (> 27) [33, 34]. Anxiety scores are categorized as normal (0–7), mild ([8–9]), moderate [10–14], and severe (> 19) [33, 34]. Stress scores are grouped into normal (0–14), mild [15–18], moderate [19–25], severe [26–33], and very severe (> 33) [38, 39]. Additionally, if a person is healthy, then a score of more than 9 for depression, 7 for anxiety, and 14 for stress typically suggests that they have an excessive number of symptoms [39]. The Iranian population can rely on the DASS-21 scale as a dependable and effective means of evaluating depression, anxiety, and stress [40]. The original DASS-21 has demonstrated good internal consistency, with Cronbach’s alpha values of 0.93 for depression, 0.89 for anxiety, and 0.91 for stress [38]. In the Persian validation study by Kakemam et al. (2022), Cronbach’s alpha coefficients were 0.93 for depression, 0.79 for anxiety, and 0.91 for stress, indicating acceptable reliability in the Iranian population [40].

Evaluation of food intake

Dietary intake was assessed with a 147-item food frequency questionnaire (FFQ), previously validated for the Iranian population [41]. Interviews were conducted by trained nutritionists, and intakes were converted to grams and analyzed using Nutritionist 4 (v7) software [42]. Nutrient composition was primarily derived from the USDA Food Composition Databases, while the Iranian Food Composition Table was used for local foods not available in the USDA database.

Vegetables were categorized into starchy and non‑starchy groups according to their relative carbohydrate and starch content, consistent with nutritional classification approaches used in epidemiological studies. Starchy vegetables included potatoes, sweet potatoes, corn, green peas, and other high‑starch vegetables commonly consumed in the Iranian diet.

Non‑starchy vegetables included leafy greens (such as lettuce, parsley, spinach, and leeks), cruciferous vegetables (e.g., broccoli, cabbage, and cauliflower), fruit vegetables (e.g., tomatoes, eggplant, and peppers), and other vegetables such as mushrooms.

All vegetable items reported in the FFQ were assigned to one of these categories. Items not explicitly listed were classified based on their nutrient composition, ensuring that no vegetable items were excluded from the analysis.

Participants were categorized into low- and high-consumption groups based on the median intake of each vegetable type.

Sleep quality

Sleep quality over the past month was evaluated using the Persian version of the Pittsburgh Sleep Quality Index (PSQI), with scores ranging from 0 to 21; scores ≥ 5 indicate poor sleep quality [43]. This Index has been verified and authorized for use in the Iranian population [44].

Circadian rhythm

Chronotype was assessed using the Morning–Evening Questionnaire (MEQ), validated for the Iranian population. Participants were classified as morning (≥ 59), intermediate [42–58], or evening types (≤ 41). This questionnaire was validated in Rajabi’s study (2019) and approved for use in the Iranian population [45].

Anthropometric measurements

Height and weight were measured using a Seca stadiometer and digital scale, respectively, with participants wearing light clothing and no shoes. BMI was categorized according to the World Health Organization (WHO) criteria as overweight (25.0–29.9 kg/m²) and obese (≥ 30.0 kg/m²). Waist‑to‑hip ratio (WHR) was calculated as waist circumference divided by hip circumference. According to WHO guidelines for women, a WHR ≥ 0.85 was considered indicative of increased cardiometabolic risk [46].

Biochemical assessments

Fasting blood samples were collected, centrifuged, and stored at − 80 °C. Serum glucose and lipid profiles were analyzed in the Endocrine and Metabolism Research Institute laboratory (Tehran University of Medical Sciences) using Randox kits (Hitachi 902) per manufacturer instructions. Triglycerides were measured using the GPO-PAP method; total cholesterol and HDL cholesterol were assessed using enzymatic endpoint and clearance assays, respectively [47].

Blood pressure

Systolic and diastolic blood pressure were measured twice on the right arm using a standard mercury sphygmomanometer, and the average of the last two readings was used. A third measurement was taken if the difference exceeded 5 mmHg [48].

Statistical analysis

Normality was assessed using the Kolmogorov–Smirnov test. Quantitative data are reported as mean ± SD and qualitative data as n (%). Independent t-tests compared demographic characteristics, anthropometrics, blood pressure, and nutrient intake between low- and high-vegetable intake groups. Binary logistic regression was used to evaluate associations between starchy and non-starchy vegetable intake and mental health, circadian rhythm, and sleep quality, reporting odds ratios (OR) and 95% confidence intervals (CI). Adjusted models included age, physical activity, and energy intake (Model 1). A fully adjusted model (Model 2) additionally included BMI, education level, marital status, socioeconomic status, job status, and total fruit intake. Multicollinearity among independent variables was assessed using variance inflation factors (VIF), and all values were below 5, indicating no significant multicollinearity. Nutrient intake comparisons between high and low vegetable intake groups were adjusted for total energy intake using ANCOVA. The study sample was homogeneous with respect to age (36.7 ± 9.1 years) and BMI (31.3 ± 4.3 kg/m²), minimizing confounding by these factors. Statistical analyses were performed using IBM SPSS Statistics 22, with p < 0.05 considered significant. Results are also presented graphically using forest plots to illustrate ORs and 95% CIs. To account for the multi-stage cluster sampling design, robust standard errors were applied to adjust for potential intra-cluster correlation by health center.

Results

Study population characteristics

The present study included 391 overweight or obese women. Table 1 shows the general characteristics of participants across categories of starchy and non-starchy vegetable consumption. Participants with higher non-starchy vegetable intake were significantly older (P < 0.001), reported greater levels of physical activity (P = 0.02), and were more likely to be married (P = 0.01) and university-educated (P = 0.05) compared with those with lower intakes. In the crude model, a higher BMI was observed in the high non-starchy vegetable intake group (P = 0.01), and this association remained significant after adjustment for age, energy intake, and physical activity (P = 0.01).

Table 1.

General characteristics of participants according to starchy and non-starchy

Quantitative	Total	Starchy vegetable				Non-Starchy vegetable
		Mean±SD				Mean±SD
		Low n: 196	High n: 195	P value	P value*	Low n:196	High n:195	P value	P value*
Demographic characteristic
Age (year)	36.7 ± 9.1	37.12+9.59	36.28+8.84	0.36	0.35	35.31+9.59	38.08+8.61	<0.001	<0.001
Physical activity	1202.1 ± 2085.3	984.29+211	1384.68+2086.	0.13	0.14	901.62+1256	1483.82+2639	0.02	0.02
Anthropometric measurements
Weight (kg)	81.3 ± 12.4	79.69+10.31	80.57+11.29	0.42	0.63	79.25+10.19	81.01+11.34	0.11	0.48
Height (cm)	161.2 ± 5.9	160.95+5.99	161.35+5.79	0.50	0.28	161.49+5.89	160.80+5.88	0.24	0.05
BMI (kg/m2)	31.3 ± 4.3	30.81+3.79	31.13+3.97	0.42	0.35	30.46+3.78	31.47+3.92	0.01	0.12
WC (cm)	31.3 ± 4.3	99.01+8.92	99.23+9.90	0.82	0.73	98.31+8.74	99.93+9.99	0.09	0.39
WHR	1.2 ± 4.5	1.40+6.52	0.93+0.05	0.31	0.15	1.40+6.52	0.94+0.05	0.32	0.58
HC	114.0 ± 9.7	112.65 + 7.03	113.32 + 8.40	0.53	0.47	112.87 + 7.87	113.26 + 7.83	0.46	0.56
Blood pressure
SBP (mmHg)	111.4 ± 14.8	111.77+12.58	111.61+14.68	0.92	0.29	110.65+12.08	112.66+15.06	0.22	0.86
DBP (mmHg)	77.6 ± 10.4	78.48+10.89	76.69+10.00	0.15	0.05	77.87+11.17	77.21+9.75	0.59	0.27
Biochemical variables
FBS (mg/dl)	87.5 ± 9.6	86.96+10.37	87.93+9.00	0.44	0.69	87.60+9.95	87.40+9.37	0.87	0.32
Chol (mg/dl)	185.3 ± 35.8	185.74+36.37	184.68+36.28	0.82	0.56	184.18+35.66	186.00+36.88	0.69	0.84
TG (mg/dl)	118.10 ± 58.88	117.67+60.05	118.81+59.77	0.88	0.94	114.76+55.61	121.48+63.33	0.38	0.45
HDL-C (mg/dl)	46.6 ± 10.9	46.27+9.90	47.24+11.58	0.48	0.35	46.66+9.70	46.94+11.80	0.84	0.46
LDL-C (mg/dl)	95.3 ± 24.1	94.72+23.80	95.29+24.59	0.85	0.96	95.38+24.17	94.72+24.30	0.83	0.92
Qualitative variables
Education
Lower diploma	53	29(54.7)	24(45.3)	0.77	0.78	19(35.8)	34(64.2)	0.05	0.05
Diploma	152	76(49.7)	76(50.3)			75(49.0)	77(51.0)
Bachelor's or higher	186	91(49.2)	95(50.8)			102(54.6)	84(45.4)
Marital situation
Single	110	53(48.2)	57(51.8)	0.63	0.68	66(60.0)	44(40.0)	0.01	0.05
Marital situation
Single	110	53(48.2)	57(51.8)	0.63	0.68	66(60.0)	44(40.0)	0.01	0.03
Job status
Employed	158	86(53.2)	78(46.8)	0.36	0.56	91(55.7)	73(44.3)	0.08	0.12
Unemployed	227	110(48.5)	117(51.5)			105(46.3)	122(53.7)
Economic status
Poor	93	45(46.6)	48(53.4)	0.12	0.13	43(50.0)	47(50.0)	0.09	0.08
Moderate	188	102(55.5)	84(44.5)			103(56.0)	85(44.0)
Good	113	49(43.9)	63(56.1)			50(43.0)	63(57.0)

Data are presented as mean ± standard deviation (SD) for quantitative variables and as number (percentage) for qualitative variables. P‑values < 0.05 were considered statistically significant. P‑values obtained from an independent t-test; * P‑values adjusted for age, total energy intake, and physical activity using multivariate analysis. P-value* was adjusted for age, energy, and physical activity by multivariate analysis. Bold values indicate statistically significant results (P < 0.05)

Dietary intake of study subjects

Table 2 presents the dietary intake of participants according to starchy and non-starchy vegetable consumption. Among individuals with higher starchy vegetable intake, the mean intake of glucose, fructose, vitamins A, C, E, B6, B9, and B12, magnesium, and total dietary fiber was significantly greater than in the low-intake group (P < 0.05). No other macronutrients or micronutrients differed significantly between starchy vegetable intake categories.

Table 2.

Dietary intakes of macronutrients and micronutrients by vegetable intake groups

Dietary intakes	Total	Starchy vegetables		P value	Non-starchy vegetables		P value
		Mean±SD			Mean±SD
		Lown: 196	Highn: 195		Lown: 196	Highn: 195
Macronutrients
Energy(kcal/day)	2633.3 ± 809.4	2328.93+713.62	2939.18+785.86	<0.001	2426.04+788.42	2841.57+778.01	<0.001
Protein (g/day)	91.3 ± 31.5	80.04+27.32	102.64+31.33	0.08	82.36+30.52	100.30+29.83	0.008
Carbohydrate (g/d)	372.5 ± 124.6	329.05+108.12	416.07+125.06	0.91	336.89+122.40	408.18+116.56	0.007
Total fat (g/d)	95.1 ± 35.2	83.87+31.50	106.46+35.09	0.83	90.142+33.92	100.16+35.77	0.008
Monosaccharaides
Glucose	19.9 ± 10.7	16.61+7.94	23.24+12.03	0.02	15.57+ 8.02	24.29 + 11.28	<0.001
Fructose	23.9 ± 12.6	20.00+ 9.80	27.76+ 13.87	0.01	18.71 + 9.63	29.05 + 13.11	<0.001
Sucrose	31.4 ± 19.9	28.04+ 18.88	34.83 + 20.36	0.69	27.05 + 18.27	35.82 + 20.54	0.02
Types of fatty acids
SAFA (g/d)	28.4 ± 11.5	25.50+10.33	31.33+11.97	0.23	27.10+11.20	29.71+11.75	<0.001
MUFA (g/d)	32.0 ± 12.9	28.13+11.60	35.89+13.03	0.50	30.84+12.83	33.18+12.92	<0.001
PUFA (g/d)	20.1 ± 9.6	17.24+8.57	22.93+9.69	0.07	19.42+9.88	20.74+9.21	0.02
TFA (mg/d)		0.001+0.00	0.003+0.00	0.98	0.001+0.00	0.002+0.00	0.11
Vitamins
A (µg/day)		591.64 + 279.87	935.13 + 439.08	<0.001	586.11 +294.79	940.69 + 424.59	<0.001
C (mg/d)	188.4 ± 116.8	155.27+89.46	221.71+130.91	0.01	144.85+110.58	232.18+106.18	<0.001
D (μg/d)	2.0 ± 1.6	1.81+1.48	2.10+1.61	0.61	1.72+1.46	2.19+1.62	0.12
E (mg/d)	17.0 ± 9.1	14.77+8.16	19.26+9.35	0.05	16.77+10.22	17.25+7.71	0.04
B1 (mg/d)	2.1 ± 0.7	1.92+0.67	2.35+0.73	0.18	1.99+0.74	2.28+0.69	0.42
B2 (mg/d)	2.3 ± 0.9	2.03+0.81	2.51+0.86	0.87	2.06+0.86	2.49+0.82	0.09
B3 (mg/d)	25.65±9.5	23.32+8.75	29.40+10.46	0.89	23.84+10.29	28.88+9.26	0.12
B6 (mg/d)	2.2 ± 0.8	1.88+0.62	2.51+0.74	<0.001	1.89+0.69	2.50+0.69	<0.001
B9 (μg/d)	694.3 ± 234.9	544.50+174.57	696.42+180.47	<0.001	551.70+184.00	689.18+176.85	<0.001
B12 (μg/d)	4.3 ± 2.5	4.08+2.47	4.60+2.47	0.05	4.01+2.38	4.67+2.53	0.84
Minerals
Phosphorous(mg/d)	1674.9 ± 563.4	1473.26+505.82	1877.56+546.38	0.08	1504.22+540.62	1846.44+534.12	<0.001
Magnesium (mg/d)	475.7 ± 171.6	409.45+145.52	542.24+170.32	<0.001	413.59+153.38	538.08+166.50	<0.001
Zinc (mg/d)	13.4 ± 4.9	11.75+4.37	15.07+4.79	0.36	12.05+4.54	14.77+4.82	0.01
Selenium (mcg/d)	126.5 ± 49.6	115.04+44.87	137.96+51.57	0.09	120.41+49.72	132.56+48.88	0.02
Calcium (mg/d)	1268.6 ± 534.7	1137.61+496.53	1400.29+540.46	0.89	1127.61+484.95	1410.34+545.90	0.01
Iron (mg/d)	25.56±20.85	23.32+16.82	29.55+23.97	0.77	23.56+18.38	29.31+22.85	0.53
Total Fiber (g/d)	47.3 ± 21.4	40.20+18.81	54.51+21.40	0.01	40.04+19.44	54.67+20.71	<0.001

MUFA is a monounsaturated fatty acid; PUFA is a polyunsaturated fatty acid; SAFA is a saturated fatty acid; TFA is a trans fatty acid. Data are presented as mean (standard deviation). Values for quantitative variables are represented as Mean±SD. P-values < 0.05 were considered significant. P-values were derived from ANCOVA models adjusted for total energy intake. Bold values indicate statistically significant results (P < 0.05)

In contrast, participants with higher non-starchy vegetable intake demonstrated significantly higher intakes of total energy, protein, carbohydrates, and total fat (all P < 0.001*). Likewise, the intake of monosaccharides (glucose, fructose, and sucrose), all major fatty acid classes (SAFA, MUFA, PUFA), and several vitamins and minerals—including vitamins A, C, E, K, B6, B9, phosphorus, magnesium, zinc, selenium, calcium, and total fiber—was markedly greater in the high-intake group (P < 0.05*). No significant differences were observed for trans fatty acids, vitamins D, B1, B2, B3, B12, or iron.

Vegetable intake in study subjects

Table 3 presents the average consumption of starchy and non-starchy vegetables among participants. The mean intake of non-starchy vegetables was substantially higher (337.99 ± 229.92 g/day) than that of starchy vegetables (70.54 ± 50.66 g/day). The median intake was 58.1 g/day for starchy vegetables and 292.27 g/day for non-starchy vegetables. Based on these median values, participants were categorized into “low” and “high” intake groups for each vegetable type, and all subsequent analyses were conducted according to these groupings

Table 3.

Daily intake of starchy and non-starchy vegetables among study participants

	Starchy vegetable	Non-Starchy vegetable
N	391	391
Mean±SD (g /d)	70.54 + 50.66	337.99 + 229.92
Median (g/d)	58.1	292.27

Values are expressed as mean ± standard deviation (SD) and median (g/day). Median values were used to classify participants into low and highintake groups for each vegetable type in subsequent analyses

The association of Starchy and Non-starchy vegetables with depression, anxiety, and stress

Figure 1 illustrates the associations between starchy and non-starchy vegetable consumption and psychological outcomes, including depression, anxiety, and stress. A significant inverse association was observed between non-starchy vegetable intake and both depression and stress in the crude and adjusted models. Participants with higher non-starchy vegetable consumption exhibited approximately 60% lower odds of depression (crude model: OR = 0.40, 95% CI = 0.17–0.92, P = 0.03; adjusted model: OR = 0.40, 95% CI = 0.20–0.79, P < 0.001) compared with those with lower intakes. Likewise, higher intakes of non‑starchy vegetables were associated with a significant reduction in stress risk (crude model: OR = 0.52, 95% CI = 0.28–0.97, P = 0.04; adjusted model: OR = 0.30, 95% CI = 0.14–0.85, P = 0.02). Regarding anxiety, there was no significant association with non-starchy vegetable intake in either the crude or adjusted models (adjusted OR = 0.65, 95% CI = 0.34–1.21, P = 0.17; fully adjusted OR = 0.78, 95% CI = 0.38–1.60, P = 0.45). Similarly, starchy vegetable intake was not significantly associated with anxiety (adjusted OR = 0.90, 95% CI = 0.50–1.62, P = 0.73; fully adjusted OR = 0.69, 95% CI = 0.32–1.45, P = 0.32).

Fig. 1.

Associations of starchy and non-starchy vegetable intake with depression, anxiety, and stress based on logistic regression models. CI, confidence interval; OR, odds ratio. Binary logistic regression models were adjusted for age, physical activity, and total energy intake. Adjust model 2: Model 1+ BMI, education level, marital status, socioeconomic status, job status, total fruit intake .The lowintake group was considered the reference category. Pvalues < 0.05 were regarded as statistically significant

The association between PSQI and circadian rhythm with starchy and non-starchy vegetables

Table 4 demonstrates the associations of starchy and non-starchy vegetable consumption with sleep quality (PSQI) and circadian rhythm among overweight and obese women. A significant inverse association was observed between non-starchy vegetable intake and poor sleep quality in both the crude and adjusted models. Participants with higher consumption of non‑starchy vegetables had 49% lower odds of poor sleep quality in the crude model (OR = 0.51, 95% CI = 0.32–0.79, P < 0.001), 59% lower odds in the adjusted model (OR = 0.41, 95% CI = 0.22–0.78, P < 0.001), and 62% lower odds in the fully adjusted model (OR = 0.38, 95% CI = 0.19–0.77, P = 0.007), compared with those with lower intakes.

Table 4.

Associations of starchy and non-starchy vegetable intake with sleep quality (PSQI) and circadian rhythm patterns based on logistic regression analysis

Variable			Starchy vegetables			Non-starchy vegetables
			OR	95%CI	P-value	OR	95%CI	P-value
PSQI	Crude	> 5 poor	0.69	0.44,1.0	0.10	0.51	0.32,0.79	<0.001
	Adjust 1		0.83	0.44,1.57	0.57	0.41	0.22,0.78	<0.001
	Adjust 2		0.74	0.38,1.46	0.39	0.38	0.19,0.77	0.007
Circadian rhythm	Crude	Morning	0.46	0.24,0.88	0.02	0.73	0.39,1.37	0.33
		Evening	1.03	0.63,1.69	0.89	1.12	0.68,1.83	0.64
	Adjust 1	Morning	0.35	0.17, 0.72	<0.001	0.71	0.37,1.39	0.32
		Evening	1.12	0.65, 1.92	0.68	1.05	0.62,1.78	0.84
	Adjust 2	Morning	0.27	0.09,0.83	0.02	0.57	0.18,1.39	0.18
		Evening	0.69	0.32,1.45	0.32	0.75	0.35,1.58	0.45

CI, confidence interval; OR, odds ratio; PSQI, Pittsburgh Sleep Quality Index. P‑values < 0.05 were considered statistically significant. Multinomial logistic regression models were adjusted for the following in model 1: age, physical activity, and total energy intake. The low‑intake group served as the reference category for both PSQI (< 5 = good sleep) and circadian rhythm (intermediate type).Adjust model 2: Model 1+ BMI, education level, marital status, socioeconomic status, job status, total fruit intake. Bold values indicate statistically significant results (P < 0.05)

For circadian rhythm, higher intake of starchy vegetables was associated with lower odds of being a morning-type individual in both the adjusted model (OR = 0.35, 95% CI = 0.17–0.72, P < 0.001) and the fully adjusted model (OR = 0.27, 95% CI = 0.09–0.83, P = 0.02).

Discussion

The present cross‑sectional study examined whether the intake of starchy and non‑starchy vegetables was associated with depression, anxiety, stress, circadian rhythm, and sleep quality among overweight and obese women. Higher non‑starchy vegetable intake was linked to lower odds of depression, stress, and poor sleep quality, whereas a higher intake of starchy vegetables was inversely associated with a reduced likelihood of morning-type chronotype (adjusted OR = 0.35, 95% CI = 0.17–0.72, P < 0.001; fully adjusted OR = 0.27, 95% CI = 0.09–0.83, P = 0.02), indicating that higher starchy vegetable consumption may shift circadian preference toward evenings. This study is among the first to simultaneously differentiate between starchy and non‑starchy vegetables in relation to these psychological and behavioral outcomes, addressing an area that has received limited attention in previous literature.

Dietary patterns rich in vegetables have long been associated with improved mental-health outcomes. Previous studies, including a recent meta-analysis of 23 studies, have shown that higher vegetable intake correlates with lower levels of depression, anxiety, and psychological distress [49]. Furthermore, vegetable intake also plays an important role in affecting sleep duration [26]. Similarly, adherence to the Mediterranean diet — characterized by abundant consumption of vegetables — was inversely associated with the odds of depression [50]. In women, intake of dark-green leafy vegetables, mainly classified as non-starchy types, was significantly linked to a lower risk of major depression [51]. Evidence from a 15-year longitudinal cohort of young women in Australia also showed that higher vegetable intake predicted a decreased risk of depressive symptoms [52]. Among the studies that specifically examined a type of vegetable, it has been shown that carrots, lettuce, cucumbers, and green leafy vegetables, especially spinach, have a better relationship with mental health [53]. The intake of vegetables, which provide essential nutrients like folate and B vitamins, particularly B12, may help alleviate symptoms of depression through improved communication [54–56]. Nonetheless, some studies have reported null associations [57], suggesting that cultural and dietary contexts may modify this relationship.

In the present study, a higher non-starchy vegetable intake was initially associated with higher BMI; however, this relationship became non-significant after adjustment for age, energy intake, and physical activity. This attenuation suggests that the crude association was likely confounded by participant characteristics, particularly age and physical activity, as individuals with higher vegetable intake were also older and more physically active. Therefore, the observed association does not appear to reflect an independent effect of vegetable intake on BMI but rather underlying lifestyle differences within the study population.

Regarding stress and anxiety, recent cross-sectional evidence among adults in Australia revealed that intakes of vegetables in diverse colors (yellow/orange, red, and green) were associated with lower perceived stress [58]. Also, individuals consuming at least three servings of vegetables daily had reduced psychological distress [59]. These results suggest that higher vegetable diversity and antioxidant intake could improve mental resilience by decreasing oxidative stress and inflammatory cytokines [60–62]. However, given the cross-sectional design of most studies, causality remains uncertain. Lifestyle factors such as physical activity, sleep, and social support could further mediate these associations.

Evidence directly linking vegetable consumption and circadian rhythm in humans remains scarce. While our study identified a significant inverse association between starchy-vegetable intake and morning-type pattern, mechanistic support is limited. Nevertheless, previous studies emphasize that meal timing and macronutrient composition — particularly carbohydrate intake — affect the synchronization of peripheral clocks [14]. This novel observation should, therefore, be further investigated under controlled experimental settings.

Our findings on sleep quality (PSQI) also concur with large epidemiological data demonstrating improved sleep duration and quality with higher consumption of fruits and vegetables. For instance, higher intake (> 2 servings/day) has been linked to 21–33% lower odds of short or long sleep [63], and increasing consumption by at least three servings improves insomnia symptoms, sleep quality, and sleep latency [64]. Long-term findings also show that vegetable-rich diets predict longer sleep duration [65] and fewer insomnia complaints with faster sleep onset [66]. Non-starchy vegetables are considered beneficial for sleep due to their nutrient density, including vitamins (such as folate and vitamin C), minerals (such as magnesium and potassium), and bioactive compounds like polyphenols and fibre, which are involved in neurotransmitter synthesis, circadian regulation, and the modulation of inflammatory pathways [67]. A recent meta-analysis confirmed that healthy dietary patterns rich in vegetables, fruits, and whole grains are significantly associated with better sleep quality and lower insomnia risk [29]. Previous studies have also shown the influence of carbohydrate composition on slow‑wave sleep (SWS) and non‑rapid eye movement (NREM) sleep [67].

Vegetables have the potential to be associated with sleep quality due to their high levels of melatonin and serotonin [68, 69]. Furthermore, vegetables are known for being a valuable source of antioxidants, which studies have shown can enhance sleep quality by diminishing oxidative stress [67, 68]. Research has indicated that while sleeping, the metabolic rate and temperature of the brain decrease compared to when one is awake. This could potentially aid in restoring the enzymes that free radicals have impacted. Consuming antioxidant-rich foods, such as vegetables, is believed to have a positive impact on sleep quality, potentially improving it [70]. Conversely, when poor sleep quality occurs, individuals may experience restlessness and emotional distress, leading to an increased intake of high-calorie foods and decreased vegetable consumption. Hence, there is a reciprocal relationship between vegetable consumption and sleep parameters [71].

The health advantages of vegetables in mitigating the risk of cardiovascular disease, stroke, and cancer are well-established [72]; our results, particularly the robust associations observed in the fully adjusted models (e.g., OR = 0.38 for sleep quality and OR = 0.27 for morning chronotype), suggest a possible association with psychological well-being and sleep health, but this should be confirmed in longitudinal studies. However, evidence distinguishing between starchy and non-starchy types remains limited. Addressing inadequate vegetable consumption remains a vital public-health priority.

This study’s strengths include being the first to simultaneously assess two sub-categories of vegetable intake in relation to psychological and sleep outcomes, employing validated questionnaires adapted for Iranian women. Yet, this study has several limitations. Its cross‑sectional design prevents causal inference, and the use of self‑reported FFQs may introduce recall and reporting biases. Residual confounding is possible despite adjustment for major lifestyle factors, and the female‑only sample limits generalizability to the broader population. Although we adjusted nutrient comparisons for total energy intake, residual confounding cannot be fully excluded. While our sample was homogeneous in age and BMI, absolute nutrient intakes may still be influenced by body size and metabolic requirements. Furthermore, the cross-sectional design precludes determining the direction of causality; reverse causality cannot be ruled out, as individuals with depression, stress, or poor sleep quality may have lower motivation or capacity to maintain healthy dietary habits, including vegetable consumption. Additionally, higher non-starchy vegetable intake may reflect an overall healthier dietary pattern and lifestyle rather than an independent effect of this vegetable subgroup alone. It is also important to acknowledge that vegetable intake is often embedded within a broader healthy lifestyle pattern. A recent study reported that dietary patterns are influenced not only by nutrition knowledge but also by dieting self-efficacy, physical activity, and other behavioral factors [73]. In line with this, our study observed that participants with higher non-starchy vegetable intake also had higher physical activity levels, higher education, and different marital status compared with those with lower intake. Therefore, the observed associations between vegetable intake and mental health or sleep quality should be interpreted cautiously, as they may partly reflect the influence of an overall healthy lifestyle rather than the independent effect of vegetable consumption alone.

Longitudinal studies are needed to establish temporal relationships. Nevertheless, the study has notable strengths, including being one of the first to differentiate between starchy and non‑starchy vegetable subgroups in relation to psychological and sleep outcomes. The use of validated tools, a relatively large sample of overweight and obese women, and a comprehensive assessment of dietary, psychological, and behavioral variables provide a novel and multidimensional perspective on this relationship.

Conclusion

In summary, our study underscores the distinct and significant role that vegetable subcategories may be related to psychological well-being and sleep quality among overweight and obese women. The robust inverse associations observed between non‑starchy vegetable intake and depression, stress, and poor sleep quality—which remained significant even after full adjustment for multiple confounders (e.g., OR = 0.38 for sleep quality)—highlight these nutrient‑dense foods as potential dietary factors that may be associated with mental health and sleep hygiene, though causal relationships cannot be established in this high‑risk population. Furthermore, the novel finding regarding the inverse relationship between starchy vegetables and the morning‑type circadian pattern suggests that the carbohydrate and glycemic profile of vegetables may interact with peripheral molecular clocks, a relationship that warrants further investigation.

Given the cross‑sectional nature of this study, which precludes causal inference, future research must move toward longitudinal and interventional designs to confirm these findings. Specifically, future studies should incorporate objective biological markers—such as salivary melatonin, cortisol, or inflammatory cytokines—to elucidate the precise mechanistic pathways involved. Additionally, we recommend that future investigations explore the “chrononutrition” aspect of vegetable consumption; identifying whether the time of day (e.g., evening vs. morning) when starchy versus non‑starchy vegetables are consumed influences metabolic and psychological outcomes could provide more tailored and effective dietary guidelines for weight management and mental resilience.

The study was supported by the Tehran University of Medical Sciences Grant (No: 97-03-161-41155). The researchers expressed their gratitude to all those who participated in the research and the Faculty of Nutrition and Dietetics at Tehran University of Medical Sciences.

Abbreviations

WHO

World health organization

WC

Waist circumference

WHR

Waist to hip ratio

HC

Hip circumference

BMI

Body mass index

FFQ

Food frequency questionnaire

PSQI

Pittsburgh sleep quality index

MEQ

The morning–evening questionnaire

IPAQ

International physical activity questionnaire

TG

Triglyceride

T-chol

Total cholesterol

HDL-C

High-density lipoprotein cholesterol

SBP

Systolic blood pressure

DBP

Diastolic blood pressure

HTN

Hypertension

CVD

Cardiovascular disease

DASS-21

Depression anxiety and stress scale

SAFA

Saturated fatty acid

MUFA

Monounsaturated fatty acid

PUFA

Polyunsaturated fatty acid

BFM

Body fat mass

FFM

Fat-free mass

BF

Body fat

SD

Standard deviation

CI

Confidence interval

SWS

Slow wave sleep

NREM

Non-rapid eye movement

Author contributions

ZA and AM contributed to the conception and design. ZR, PJ, and RR contributed to writing and experimental procedures. FA handled data and statistical analysis. AM and KhM supervised the entire project. All authors participated in editing and approving the final version of the paper and contributed to the finalization of the manuscript. They also read and approved the final manuscript. This demonstrates a collaborative effort in the research process.

Funding

This study was supported by the Tehran University of Medical Sciences (TUMS) (Grant No: 97-03-161-41155).

Data availability

The findings of this study are supported by data that can be obtained from Dr. Khadijeh Mirzaei, although there are restrictions on their availability. These data were used under license for the current study and are not accessible to the public. However, the authors can provide the data upon reasonable request and with the permission of Dr. Khadijah Mirzaei.

Declarations

Ethics approval and consent to participate

The authors and the Ethics Committee of Tehran University of Medical Sciences (IR.TUMS.VCR.REC.1398.692) have received approval for the method and protocol of the current study. Additionally, written informed consent has been obtained from all participants.

Consent for publication

All authors collaborated in editing this article, and the final version was approved for submission. Furthermore, they actively participated in writing the manuscript and giving their approval to the final draft.

Competing interests

The authors declare no competing interests.