Anxiety, depression and psychological distress are recognised as leading causes of disability, and according to the 2017 Global Burden of Disease report, depression has impacted 264 million people worldwide(1). Notably, international and national statistics consistently report a higher prevalence of depression among women compared with men(2). In Iran, approximately 20 % of the adult population experience anxiety and depression, with a higher prevalence among women compared with men(Reference Naghavi, Abolhassani and Pourmalek3). Recent reviews have highlighted strong connections between mental and physical health, particularly with CVD(Reference Nielsen, Banner and Jensen4), cancer(Reference Wang, Li and Shi5), all-cause mortality(Reference Machado, Veronese and Sanches6) and reduced life expectancy.
Research has consistently shown that women face unique challenges and vulnerabilities that can potentially impact their mental well-being. Recognising this sex-specific burden, the WHO has identified women’s health as a critical priority, highlighting the need for focused research in this area due to the historically limited and sometimes unreliable body of research dedicated to women’s health(7). Furthermore, women’s mental health during the reproductive years holds particular significance, as it can have far-reaching implications, including for fertility outcomes.
Diet has gained attention as a modifiable risk factor for mental health problems. A recent review emphasised the importance of EPA, DHA, vitamin E, Mg and folic in psychiatric disorders(Reference Muscaritoli8). Furthermore, previous studies have observed associations between higher consumption of fruits, vegetables, fish, legumes and nuts with lower risk of psychological disorders(Reference Anjom-Shoae, Sadeghi and Keshteli9–Reference Walsh, Lee and Best16). Conversely, a recent meta-analysis demonstrated a positive association between intake of red and processed meat and depression(Reference Nucci, Fatigoni and Amerio17). Considering the synergistic effects of nutrients and foods, investigating diet-disease associations from a holistic perspective (examining the entire diet) is a desirable approach.
Previous studies have explored the relationship between mental health and various diet quality scores, including the Mediterranean diet score(Reference Sadeghi, Keshteli and Afshar18), the Healthy Eating Index(Reference Wang, Zhao and Nie19), the dietary phytochemical index(Reference Darooghegi Mofrad, Siassi and Guilani20), the food quality score(Reference Darooghegi Mofrad, Siassi and Guilani21), dietary total antioxidant capacity(Reference Daneshzad, Keshavarz and Qorbani22) and the recommended food score(Reference Lee, Kim and Park23). These indices primarily recommend increased consumption of plant-based foods while limiting intake of animal foods(Reference Sadeghi, Keshteli and Afshar18–Reference Daneshzad, Keshavarz and Qorbani22).
In recent years, due to the growing prevalence of diet-related diseases and the bidirectional impact of food systems on climate change, researchers have shown increased interest in sustainable diets(Reference Myers, Smith and Guth24–Reference Tepper, Geva and Shahar26). Sustainable diets are characterised by their affordability and cultural acceptability, health-promoting effects and lower environmental impacts(Reference Burlingame and Dernini27). They primarily consist of plant-based foods that provide protection against both adverse health outcomes and environmental consequences(Reference Tepper, Geva and Shahar26). Notably, Willett et al. recently developed specific guidelines known as the EAT-Lancet recommendations for a healthy diet from sustainable food systems(Reference Willett, Rockström and Loken25). These recommendations emphasise high consumption of whole grains, vegetables, fruits, legumes, nuts and unsaturated oils, moderate intake of poultry and seafood and limited or no consumption of refined grains, starchy vegetables, red and processed meat and added sugar(Reference Willett, Rockström and Loken25). Several indices have been proposed to combine the environmental and health aspects of diet(Reference Seconda, Baudry and Pointereau28,Reference Trijsburg, Talsma and Crispim29) . The (World Index for Sustainability and Health) WISH was established based on the EAT-Lancet recommendations(Reference Willett, Rockström and Loken25) and encompasses both diet quality and environmental sustainability in a single scoring system(Reference Trijsburg, Talsma and Crispim29). However, to date, the association between WISH and mental disorders has not been investigated. Therefore, the present observational study aims to evaluate the potential relationship between WISH and mental health outcomes, including depression, anxiety and psychological distress in a population of Iranian women.
Methods and materials
Study population
This cross-sectional study was conducted with women who attended ten healthcare centres in the south of Tehran from September 2017 to September 2018. The prevalence of mental disorders was used to estimate the sample size, with P = 26 %(Reference Noorbala, Bagheri Yazdi and Yasamy30); α = 0·05; d = 4·12, using the formula:
$$\;N = {{\left[ {{{({z_{1 - { \propto \over 2}}})}^2}P\left( {1 - P} \right)} \right]} \over {{d^2}}}$$
. The following inclusion criteria were considered: (1) being an Iranian woman aged 20–50 years, (2) having no history of chronic disease (CVD, cancer, diabetes, thyroid disease, liver disease, pulmonary disease, multiple sclerosis, epilepsy or kidney dysfunction) or psychological disease requiring antipsychotic medications and (3) not currently following a specific diet. Pregnant, lactating and menopausal women were excluded. Participants with energy intake higher than 3500 or lower than 500 kcal/d (n 31) and those with missing psychological profile data on the Depression and Anxiety Stress Scale-21 (n 3) were excluded from the analysis. Ultimately, 479 individuals were eligible for the present study.
Dietary intake assessment
Participants’ dietary intakes were collected using a validated semi-quantitative 168-item Iranian FFQ, administered through face-to-face interviews(Reference Mirmiran, Esfahani and Mehrabi31). A trained dietitian asked about the frequency of consumption on a daily, weekly or monthly basis, as well as the usual serving size of food items. Then, a guideline for household measures(Reference Ghaffarpour, Houshiar-Rad and Kianfar32) was used to convert serving sizes to g/d. Finally, energy and main nutrients were computed using a modified version of Nutritionist IV software for Iranian foods (version 7.0; N-Squared Computing, Salem).
World Index for Sustainability and Health (WISH) calculation
The WISH was determined based on a method developed by Trijsburg et al. (Reference Trijsburg, Talsma and Crispim29), which scores both diet healthiness and environmental impact. Detailed information about the construction and validation of WISH has been provided elsewhere(Reference Trijsburg, Talsma and Crispim29). WISH scores have been applied and calculated to other lower- and middle-income countries (e.g. for women in rural East Africa), supporting the index’s broader applicability(Reference Keding, Sarfo and Pawelzik33). The WISH considered thirteen food groups, including vegetables, fruits, whole grains, fish, red meat, dairy products, chicken, eggs, nuts, legumes, saturated oils, unsaturated oils and added sugars. Food groups were categorised based on their health-related (protective, neutral or limited) and environmental (low, medium or high) impacts. This score includes four sub-scores: healthy (vegetables, fruits, whole grains, fish, dairy products, egg, chicken, nuts, legumes, unsaturated oils), less healthy (red meat, saturated oils, added sugar), low environmental impact (whole grains, vegetables, fruits, legumes, unsaturated oils, added sugars) and high environmental impact (red meat, dairy products, egg, chicken, nuts, fish, saturated oils). For the healthy and low environmental sub-scores, a value of 0–10 indicates the lowest to the highest consumption of the components. In the case of the less healthy and high environmental sub-score, higher values correspond to less consumption of the components. Finally, a higher total score indicates a healthier or more environmentally friendly diet.
Psychological profile assessment
The Depression Anxiety Stress Scale-21 (DASS-21), a validated self-reported questionnaire, consists of seven items in three subscales, to assess psychological distress, depression and anxiety. Answers are given on a four-point Likert scale ranging from 0 (never) to 3 (always). Scores on each subscale range from 0 to 21. The Iranian version of the DASS-21, which has been validated and deemed reliable, was used in this study(Reference Sahebi, Asghari and Salari34). Since the original scale of DASS is based on forty-two questions, DASS-21 final scores in each subscale were doubled. Finally, depression, anxiety and psychological distress were defined as scores of ≥ 10, ≥ 8 and ≥ 15, respectively. The Cronbach’s alpha indices for the subscales were as follows: depression = 0·81, anxiety = 0·74, and distress = 0·78.
Anthropometric and socio-demographic assessment
Body weight was measured using a digital scale (SECA, Hamburg, Germany) while participants wore minimal clothing and no shoes. Height was measured using an inflexible measuring rod with a precision of 0·1 cm. BMI was calculated as BMI = weight (kg)/height2 (m). Demographic information including age, marital status (married/single), smoking (yes/no), socio-economic status (SES)(Reference Shafiei, Yazdani and Jadidfard35) (education, family size, employment status and occupation, homeownership, home equipment diversity, number of rooms and domestic/international travel), education, body satisfaction (yes/no), medication or vitamin supplement usage (yes/no), history of chronic disease (cancer, diabetes, CVD, liver, pulmonary, kidney, thyroid diseases, hypertension, multiple sclerosis, epilepsy) and family history of these diseases (yes/no) were collected using an interview-based questionnaire. Physical activity level was determined by having participants directly log the average time and total duration that they devoted to various physical activities over a 24-h period. Then, the physical activity level was computed as metabolic equivalent minutes per week (MET/min/week)(Reference Ainsworth, Haskell and Whitt36). Finally, four categories were used to indicate the level of physical activity (light, moderate, strong and intense).
Statistical analysis
The distribution of variables was analysed using the Kolmogorov–Smirnov test. Continuous variables were presented as mean ± sd, while categorical variables were presented as frequencies (n) and percentages (%). X 2 and one-way ANOVA tests were used to compare the general characteristics of participants across the tertiles of WISH sub-scores. Study participants’ dietary intakes across tertiles of WISH scores were compared using ANCOVA adjusted for energy intake. Binary logistic regression analysis was also applied to assess the relationship between less healthy, healthy, low environmental impact and high environmental impact scores and odds of having each of the three psychological profiles (anxiety, depression and psychological distress) by including age, energy, BMI, marital status, education, family history of chronic disease, sleep duration, duration of time spent outside, prescription medications, vitamin supplementation, body satisfaction, SES and physical activity in the multivariable-adjusted model. Covariates were selected based on clinical knowledge and a comprehensive literature review(Reference Darooghegi Mofrad, Siassi and Guilani20,Reference Darooghegi Mofrad, Siassi and Guilani21,Reference Zamani, Zeinalabedini and Nasli Esfahani37) and baseline variables associated with having each of the three psychological profiles, as illustrated in online Supplementary Fig. 1 using a directed acyclic graph. Variables that met both the statistical criteria (favourable AIC/BIC values, low multicollinearity) and the practical significance criterion (10–15 % effect size change) were selected as confounders for inclusion in our multivariable logistic regression models. Statistical analyses were performed with SPSS statistical package software (SPSS Inc. version 27), and P < 0·05 was considered significant.
Results
Table 1 presents the characteristics of women categorised according to WISH sub-scores. The mean age, weight, BMI and physical activity of the participants were 31·9 (7·7) years, 64·5 (12·0) kg, 24·5 (4·5) kg/m2 and 39·9 (6·8) MET/min/week, respectively. We observed that higher healthy, less healthy, low environmental and high environmental scores were associated with lower odds of depression and psychological distress. Additionally, women with higher healthy and low environmental and lower less healthy sub-scores reported lower levels of anxiety. No significant differences in SES, education status, supplement usage, medication use, family history of chronic disease and body satisfaction were observed.
Table 1. General characteristic of participants across the tertiles of WISH sub-scores
WISH; World Index for Sustainability and Health.
Values are means ± sd; one-way ANOVA for continuous variables and x 2 test for categorical variables were used to generate P values; MET-h, metabolic equivalent task hours.
The energy-adjusted dietary intakes of participants across tertiles of healthy, less healthy, low environmental and high environmental impacts sub-scores are presented in Table 2. Higher intakes of carbohydrate, fibre, PUFA, vitamin A, vitamin B1, vitamin B6, vitamin C, Mg, vegetables, fruit, fish, legumes, nuts and unsaturated oil and lower intakes of cholesterol, vitamin B2, vitamin D, Ca, eggs, meat and dairy products were associated with third tertiles of the healthy sub-score. Women scoring in the highest less healthy tertile had lower consumption of energy, protein, carbohydrates, fat, fibre, MUFA, vitamin B1, vitamin B2, vitamin B6, vitamin B9, Ca, Mg, Zn, meat, fruit, dairy product, saturated oil and sugar but higher intakes of vitamin A, vegetables, chicken and unsaturated oil compared with those in the lowest tertile. Women in the third tertile of low environmental impact reported significantly greater intake of energy, protein, fibre, vitamin A, vitamin B2, vitamin B6, vitamin B9, vitamin C, Ca, Mg, vegetables, fruit, dairy products, meat, legumes and unsaturated oil compared with participants in the first tertile. However, they showed lower intakes of sugar. Participants in the top environmental tertile had lower intakes of energy, protein, carbohydrate, fat, fibre, MUFA, cholesterol, vitamin D, vitamin B1, vitamin B2, vitamin B6, Ca, whole grains, fruits, egg, dairy products, meat, chicken and saturated oil compared with those in the bottom environmental tertile. Women in the lowest tertile consumed less fish and nuts.
Table 2. Energy-adjusted dietary intakes of study participants across the tertiles of WISH sub-scores
WISH: World Index for Sustainability and Health.
Values are mean ± se. All values are adjusted for energy intake using ANCOVA.
Table 3 shows adjusted OR for depression, anxiety and stress comparing tertiles of healthy, less healthy, low and high environmental scores. The healthy score was inversely associated with the odds of depression (OR 0·40; 95 % CI 0·24, 0·67; P = 0·001), anxiety (OR 0·45; 95 % CI 0·23, 0·87; P = 0·023) and distress (OR 0·46; 95 % CI 0·28, 0·77; P = 0·003). The less healthy score was positively associated with distress (OR 0·57; 95 % CI 0·36, 0·90; P = 0·016), depression (OR 0·51; 95 % CI 0·32, 0·89; P = 0·004) and anxiety (OR 0·44; 95 % CI 0·25, 0·78; P = 0·004). In addition, scores for low environmental impact were inversely associated with depression (OR 0·32; 95 % CI 0·19, 0·54; P < 0·001), distress (OR 0·30; 95 % CI 0·17, 0·51; P < 0·001) and anxiety (OR 0·38; 95 % CI 0·18, 0·76; P = 0·012). An inverse association was found between scores indicating high environmental impact and depression (OR 0·57; 95 % CI 0·33, 0·96; P = 0·031). However, we failed to detect a significant association between high environmental score with anxiety (OR 0·56; 95 % CI 0·29, 1·08; P = 0·090) and distress (OR 0·62; 95 % CI 0·37, 1·06; P = 0·080).
Table 3. Multiple-adjusted OR and 95 % CI across the tertiles of WISH sub-scores
WISH: World Index for Sustainability and Health.
These values are OR (95 % CI).
* Logistic regression model included age, energy, BMI, marriage, education, family history of chronic disease, body satisfaction, SES, physical activity and smoking.
Discussion
In this cross-sectional study, we investigated the mental health of Iranian women in relation to WISH sub-scores. Our results revealed that healthier scores were correlated with lower odds of depression, anxiety and psychological distress. In addition, we found an inverse correlation between less healthy scores and psychological disorders. Considering environmental dietary indicators, a negative association was seen between the low environmental impact sub-score and depression, anxiety and psychological distress. However, the high environmental impact sub-score was not related to the odds of mental disorders, except for depression. To the best of our knowledge, this is the first study to assess WISH scores in relation to depression, anxiety and psychological distress.
The present study was conducted among a sample of Iranian adult women. Findings indicated that Iranian women consume vegetables, fruits, dairy products, fish, chickens, eggs, nuts, legumes and unsaturated oils within the EAT-Lancet guidelines suggested range. Nevertheless, due to the moderate environmental impact of dairy products, their consumption could be slightly decreased. In low- to middle-income countries, the majority of energy intake comes from carbohydrates(Reference Seidelmann, Claggett and Cheng38). Based on the results of the present study, the average amount of whole grains consumed by this population is considerably lower than the recommended range. On the other hand, the consumption of less healthy food groups, including red meats, saturated oils and added sugar, was relatively higher than the recommended values and should be decreased. In summary, our study population could improve their current WISH score by consuming higher amounts of whole grains and fish while limiting less healthy food intake. Notwithstanding the average amounts of food consumed by our population, we should be cautious in generalising these intake values to the whole Iranian female population, as adherence to the EAT-Lancet recommendations can be relatively expensive and may differ across individuals with various SES(Reference Hirvonen, Bai and Headey39).
Vegetables, legumes, whole grains, fruits and unsaturated oils are the components of both the healthy and low environmental impact sub-scores(Reference Trijsburg, Talsma and Crispim29). In addition, fish, chicken, eggs, dairy products and nuts are known to be healthy foods(Reference Trijsburg, Talsma and Crispim29). We found an inverse association between the healthy sub-score of WISH and mental disorders. Consistent with our findings, a previous meta-analysis of observational studies found fruits, vegetables and fish, intake to be inversely linked with psychological disorders within the general population(Reference Grosso, Micek and Marventano11,Reference Saghafian, Malmir and Saneei12) . Furthermore, another cross-sectional study among 24 776 Chinese participants revealed that consumption of whole grains ≥ 2 times/week was associated with 32 and 24 % lower odds of depression in males and females, respectively(Reference Wu, Zhang and Meng40). Similarly, another cross-sectional study of 3172 Iranian adults also reported a negative association between whole-grain consumption and odds of anxiety in women, while a positive association was observed for refined grains and anxiety and depression(Reference Sadeghi, Hassanzadeh-Keshteli and Afshar41). A review of thirteen studies investigating dairy products in relation to mental disorders found inconsistent results(Reference Hockey, McGuinness and Marx42). However, because of methodological differences and considering various sub-types of dairy products in the studies included, conclusions could not be made, and further studies were recommended(Reference Hockey, McGuinness and Marx42).
We found an inverse association between the low environmental impact sub-score and odds of depression, anxiety and psychological distress. Plant-based food groups as well as added sugars are considered to have less impact on the environment. We previously mentioned results from earlier studies that had investigated the relation between specific plant-based food groups and mental health(Reference Grosso, Micek and Marventano11,Reference Saghafian, Malmir and Saneei12) . Added sugars were also included in the less environmental impact sub-score(Reference Trijsburg, Talsma and Crispim29). Despite categorisation as a low environmental impact food, added sugars were found to be an unhealthy risk factor for mental health conditions. A meta-analysis by Hu et al. (Reference Hu, Cheng and Jiang43) revealed that intake of sugar-sweetened beverages could be modestly related to depression (OR = 1·31). Also, a cohort study of 935 Japanese adults reported that higher soft drink consumption was positively associated with the risk of depression(Reference Kashino, Kochi and Imamura44). Additionally, a study among 4741 adults in South Australia concluded there was a direct link between consuming soft drinks and psychological disorders(Reference Shi, Taylor and Wittert45). Similarly, sugar and sweeteners were positively associated with anxiety(Reference Hoerr, Fogel and Van Voorhees46). Wattick et al. demonstrated a similar relation between added sugar intake and anxiety in 1959 college students(Reference Wattick, Hagedorn and Olfert47). Nevertheless, it is worth mentioning that added sugar is one of the components of the low environmental impact sub-score, and other healthy components such as vegetables, fruits and whole grains might counteract the deleterious effects of added sugars on mental health. Furthermore, EAT-Lancet guidelines recommended caution associated with added sugar intake because of the possible health concerns, suggesting a cut-off value of 31 g/d(Reference Willett, Rockström and Loken25,Reference Trijsburg, Talsma and Crispim29) .
We found that having a less healthy WISH score on any of the sub-scales was related to lower odds of psychological disorders. Higher scores indicate healthier diets with lower consumption of red meats, saturated fats and added sugars. In our earlier report, Darooghegi et al. found that women with the highest levels of red meat intake had substantially increased odds of anxiety, depression and psychological distress(Reference Darooghegi Mofrad, Mozaffari and Sheikhi48). A study by Kouvari et al. on 1514 men and 1528 women revealed that moderate (second tertile) consumption of total meat and red meat was negatively associated with odds of depression indicating a U-shaped association(Reference Kouvari, Panagiotakos and Chrysohoou49). Also, Kazemi et al. found an increased chance of depression among adults who have higher intakes of red meat, especially among people of normal weight and males(Reference Kazemi, Keshteli and Saneei50). Olivan-Blazquez et al. concluded that adults who consume more than 1 serving/d of red meat had a higher likelihood of depressive symptoms(Reference Oliván-Blázquez, Aguilar-Latorre and Motrico51). Also, a meta-analysis found that red and processed meat intake could be positively linked to high levels of depression or depression(Reference Nucci, Fatigoni and Amerio17). Furthermore, several animal studies revealed that saturated fatty acids might be a responsible nutrient associated with anxiety(Reference Meichtry, Poetini and Dahleh52–Reference Sivanathan, Thavartnam and Arif54). Despite the deleterious effect of animal products on health as well as the environment, some important nutrients including Fe, Ca, Zn and vitamin B12 are mainly provided by animal-based foods(Reference Magkos, Tetens and Bügel55). Strategies including supplementation, considering meat alternatives and promotion of nutrient absorption from plant-based foods should be considered to overcome the possible deficiencies(Reference Bastian, Buro and Palmer-Keenan56).
Dietary intake of individual nutrients and food groups is an important approach for exploring diet-disease associations. However, our dietary intakes are a combination of various foods with different nutritional values, and specific nutrients might interact in the human body. Therefore, considering dietary intake holistically could be a better option for investigating diet in relation to disease. In agreement with our findings, earlier meta-analyses reported an inverse association between the Mediterranean diet and Healthy Eating Index, which highly consists of plant-based foods as well as unsaturated fats and depression(Reference Lassale, Batty and Baghdadli57,Reference Shafiei, Salari-Moghaddam and Larijani58) . Furthermore, a cross-sectional study on Iranian adults found an inverse association between the Mediterranean diet and anxiety and psychological distress(Reference Sadeghi, Keshteli and Afshar18). Also, a cross-sectional study among Australian vegans and vegetarians found a protective relation between a plant-based diet and depression(Reference Lee, Eather and Best59). Similarly, our previous report using the same data revealed an inverse relation between overall and healthful plant-based dietary patterns and anxiety, depression and psychological distress(Reference Zamani, Daneshzad and Siassi60). However, higher scores for the unhealthful plant-based dietary pattern have been shown to be related to increased odds of depression by 91 %(Reference Zamani, Daneshzad and Siassi60). Another cross-sectional study among diabetic women in Iran found that the unhealthful plant-based diet was a predictor of increased depression, anxiety and stress, but no significant relation was found between the healthful plant-based diet and mental disorders(Reference Daneshzad, Keshavarz and Qorbani61). On the other hand, a recent meta-analysis did not find a significant relation between vegetarian diets and the risk of anxiety and depression(Reference Askari, Daneshzad and Darooghegi Mofrad62). It is important to note that discrepancies among studies may be due to different study designs and populations. Most prior studies on the relation between plant-based dietary patterns and mental disorders have been cross-sectional; thus, more prospective studies are required to rule out the possibility of reverse causation.
There are several proposed mechanisms by which healthy and low environmental impact sub-scores may be related to mental health. These sub-scores mainly reflect plant-based foods (providing considerable amounts of fibre, Mg and B vitamins), which have been found to be protective for mental disorders(Reference Skarupski, Tangney and Li63–Reference Derom, Sayón-Orea and Martínez-Ortega65). Furthermore, these sub-scores recommend unsaturated fat consumption, which has been inversely associated with depression(Reference Sánchez-Villegas, Verberne and De Irala66). The low glycaemic index of plant-based diets could be another possible mechanism. Adhering to a low glycaemic index diet could reduce insulin resistance, which has been found to be negatively associated with mental health(Reference Kan, Silva and Golden67,Reference Haghighatdoost, Azadbakht and Keshteli68) . Also, inflammation is a key factor in the pathophysiology of psychological disorders(Reference Pariante69), and adherence to plant-based diets could be an important approach for controlling inflammation(Reference Medawar, Huhn and Villringer70,Reference Bonaccio, Pounis and Cerletti71) .
In the present study, we considered possible confounding of several factors in the association between WISH score and mental disorders. These confounders included age, energy, BMI, marital status, educational level, family history of chronic disease, body satisfaction, SES, physical activity and smoking. The associations we observed were mostly unchanged after adjusting for these confounders. It should be acknowledged that residual confounding may still exist due to errors in classifying participants based on the confounding variables or due to errors in measuring confounders. Also, there may be additional confounders that we did not collect data on. For instance, sleep and psychological disorders share common causes and also have bidirectional associations (e.g. sleep disturbances have been found to be a strong causal factor for mental disorders(Reference Freeman, Sheaves and Waite72)). Also, abundant evidence indicates a potential role of diet in sleep health(Reference St-Onge, Mikic and Pietrolungo73). Given the points mentioned above, the relation between diet and mental disorders in our study might still be confounded by some mismeasured or unmeasured factors, which should be considered in the interpretation of the results. In addition, reverse causality is an important phenomenon, and the actual associations might be in the reverse direction of which we hypothesised. In this scenario, individuals with depression, anxiety or stress might alter their dietary intakes in response to their specific mental health conditions. These dietary behaviours (e.g. eating sweet foods) may result from an effort to enhance their mood, poor appetite or lack of motivation to buy or prepare healthy foods(Reference van der Pols74,Reference Northstone, Joinson and Emmett75) . For instance, a longitudinal study observed that depressive symptoms were related to changes in vegetables, meats and dairy products intake,(Reference Elstgeest, Visser and Penninx76) suggesting that a bidirectional relationship could exist between diet and depression. Such studies in which the relation between diet and mental disorders is explored in both directions are scarce and need to be prioritised in future research. To control for bias from reverse causation, we did not include participants who had used antidepressant drugs as they tend to have more severe mental disorders and are more likely to alter their dietary intakes due to their condition. Notwithstanding, a bidirectional relation may exist between dietary intakes and mental disorders(Reference Elstgeest, Visser and Penninx76), and reverse causality cannot be ruled out in our study. It is important to consider this point in interpreting our findings. Further prospective cohort studies are required to understand reverse causation.
There are some strengths of our study. It is the first study to examine the association between WISH scores and mental health conditions. Furthermore, we used validated questionnaires for dietary intake, mental health and assessment of physical activity. Also, although residual confounding might still remain, several potential confounders were taken into account in our analyses. For the interpretation of the results, several limitations were also important. First, the cross-sectional nature of our study limited our ability to infer causal relationships. Therefore, prospective studies are needed to further explore the relationships between diet and mental disorders. Second, despite using validated DASS-21 and FFQ questionnaires, they are subjected to under- or over-reporting, leading to potential participant misclassification. Third, despite being a universal score, the WISH has not been specifically validated among females or among Iranians in particular. Therefore, we recommend that future studies examine the reliability and validity of this index among different populations to assess its robustness. Finally, caution should be taken in generalising our findings to other populations including males, older adults and those with other health conditions.
In conclusion, adherence to a healthier diet with less environmental impact was associated with better psychological health among women. Due to the worsening trends in our environment and the increasing prevalence of psychological disorders, more prospective studies that include both sexes are warranted to understand the causality of the observed associations and explore potential mechanisms.
Acknowledgements
The authors would like to gratitude to the study participants.
This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.
A. J. contributed to the conception, design, search and statistical analyses. A. J. and K. L. drafted the manuscript. L. A., H. M., B. Z., M. D. M. and A. S. contributed to design and data interpretation. P. J. S. edited and commented on the manuscript. L. A. supervised the study critically. All authors have read and approved the final manuscript.
The authors declare that they have no competing interests.
All participants were provided written consent based on the guidelines of the Declaration of Helsinki. The study protocol (98-01-161-42024) was approved by the Medical Research Ethics Committee of Tehran University of Medical Sciences.
The datasets used and/or analysed during the present study are available from the corresponding author upon reasonable request.
Supplementary material
For supplementary material/s referred to in this article, please visit https://doi.org/10.1017/S0007114524000771.
