Background

Psychological conditions such as bipolar affective disorder, eating disorders, anxiety disorders, and depressive disorders impose a considerable burden across the international community, adversely affecting quality of life [1, 2]. Psychological problems including depression, stress, and anxiety, also arise in association with some non-communicable diseases including cardiovascular disease (CVD), stroke, and cancer [3]. All of these mental health problems have adverse effects on health status, quality of life, and ability to work [4].

Genetics, socioeconomic status, exercise habits, diet, and nutritional status, are understood to be key contributors to the development of emotional or behavioral problems [5]. Food-mood relationships underpin well-known pathways, suggesting that unhealthy eating habits and poor nutritional status are correlated with various mental health problems and behavioral disturbances in adults [6]. This infers that mood and psychological health may be influenced by nutritional habits [7].

The world-wide consumption of junk foods, which include ultra-processed foods, fast foods, unhealthy snacks, and sugar-sweetened beverages, is increasing. The hallmarks of junk foods are that they have high levels of energy, fat, sugar, and salt, accompanied by low levels of micronutrients, fiber, and other bioactive compounds [8]. The low nutritional value of junk foods can alter inflammatory pathways, leading to an increase in biomarkers for oxidative stress and inflammation, which contribute to biological changes associated with mental health disorders. In vitro studies have demonstrated that junk food consumption can negatively affect the brain and mental health [9, 10].

However, the findings of epidemiological studies are inconsistent. Some studies showed the significant association between junk foods consumption and mental health disorders. However, other studies did not mention any relationship [4, 11, 12]. The aim of this study is to examine the relationship between junk food consumption and mental health disorders in adults by conducting a systematic review and meta-analysis of published studies to date.

Methods

The current systematic review and meta-analysis study was conducted according to the PRISMA 2020 statement (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [13, 14], included studies assessing the relationship between junk food consumption and mental health in adults.

Search strategy

A systematic literature search was conducted in PubMed/Medline, ISI Web of Science, Scopus, Cochrane, Google Scholar, and EMBASE up to July 2023. The following keywords were used in this search: “sweetened drink*” OR “sweetened beverage*” OR snack* OR “processed food*” OR “junk food*” OR “soft drinks” OR “sugared beverages” OR “fried foods” OR “instant foods” OR sweets for junk food consumption and “mental health” OR depression OR stress OR anxiety OR “sleep dissatisfaction” OR “sleep disorders” OR happiness OR wellbeing for mental health status. In PubMed, keywords were searched through [tiab] and [MeSH] tags. Articles were required to be written in English language; there was no limitation regarding the year of publication. The reference lists of included papers were also examined to avoid missing other published data.

Inclusion and exclusion criteria

Two investigators independently screened the articles retrieved during the literature search. Publications that fulfilled the following criteria were eligible for inclusion: (1) observational studies that were conducted in adults (cohort, case-control, cross-sectional); and (2) studies that examined the relationship between junk food consumption and mental health status. We excluded letters, comments, reviews, meta-analyses, ecological, in vitro, and pre-clinical studies, as well as duplicate studies.

Data extraction

For each eligible study, the following information was extracted: first author, year of publication, study design, country, age range, gender, sample size, type of junk food, dietary assessment tool, mental health parameters, mental health assessment tool, study quality score, effect sizes and measures, and covariates.

It should be noted that in the present study, junk food intake was considered using four categories: (i) sweet drinks (fruit-flavored drinks, sweetened coffee, fruit juice drinks, sugared coffee and tea, energy drinks, cola drinks, beverages, soft drinks, lemonade, and soda), (ii) sweet snacks (total sugars, added sugars, sweetened desserts, fatty/sweet products, ice cream, chocolate, artificial sweeteners, sweet snacks, dessert, sauces and dressings, candy, patterns of consumption of sweet, high fat and sugary foods, biscuits and pastries, cakes, pie/cookies, and baked goods), (iii) snacks (including snacks, sauces/added fats, fast food, fast-food pattern, western diet pattern, snacking and convenience pattern, fried foods, fried potato, crisps, salty snacks, convenience pattern, instant foods), and (iv) total junk foods (all types of junk foods).

Quality assessment of studies

The quality of the included studies was examined using the Newcastle-Ottawa Scale (NOS) [15, 16]. The NOS assigns a maximum of 9 points to each study: 4 for selection, 2 for comparability, and 3 for assessment of outcomes (for cohort study) or exposures (for case-control study).

The maximum score for cohort and case-control studies were 9 and for cross-sectional studies were 7. In the current analysis, the quality of studies is defined good if the studies get 3 or 4 stars in the selection domain AND 1 or 2 stars in the comparability domain AND 2 or 3 stars in the outcome/exposure domain. Besides, fair quality is defined as 2 stars in the selection domain AND 1 or 2 stars in the comparability domain AND 2 or 3 stars in the outcome/exposure domain and finally, poor quality is defined for 0 or 1 star in the selection domain OR 0 star in the comparability domain OR 0 or 1 star in the outcome/exposure domain.

All steps including searching, article screening, data extraction, and quality assessment of articles were independently performed by two investigators. Disagreements between the two investigators were resolved by discussion to reach consensus.

Statistical analysis

The results of the current quantitative synthesis are presented as hazard ratios (HRs) or odds ratios (ORs) and 95% confidence intervals (95% CI). STATA version 14.0 (StataCorp, College Station, TX) software was used to perform the meta-analysis. We conducted meta-analysis whenever at least two studies investigated similar associations between junk food consumption and mental health problems.

I2 statistic and chi-square-based Q-test were used for the assessment of heterogeneity. In the current study, a lack of heterogeneity was inferred when the p-value of chi-square-based Q-test exceeded 0.10. Fixed models were used to pool HRs and ORs when the heterogeneity p-value was higher than 0.10. Random models were used to pool the ORs whenever the heterogeneity p-value was equal to or less than 0.10, followed by Galbraith analysis and sensitivity analysis. Subgroup analysis was also conducted to identify the source of heterogeneity. Publication bias was measured using Begg’s test or Egger’s test and considered substantial whenever the resulting p-value was < 0.1.

Results

Systematic search results

The flow diagram for the process of study selection is shown in the PRISMA flowchart (Fig. 1). Based on the initial search, we found 1745 papers. After removal of duplicate documents and title and abstract screening, 69 articles remained for more detailed assessment. Full texts of these papers were reviewed carefully by three researchers, with 17 articles satisfying the eligibility requirements for inclusion in the systematic review and meta-analysis.

Fig. 1
figure 1

The PRISMA flowchart for the process of study selection

Full size image

Characteristics of the included studies

Seventeen studies evaluating a total of 159,885 participants were included in our quantitative synthesis. A considerable number of participants were female, with seven articles restricted to female participants. Most of the included studies were cross-sectional (58.82%), with the remaining seven (47.05%) being cohort studies. It should be noted that Reinks et al. (2013) presented both cross-sectional and longitudinal data. Reinks et al. (2013) and ten other papers (64.70% in total) assessed depression as an outcome. Nine (52.94%) of studies assessed anxiety or stress as outcomes. In terms of dietary exposures, various types of junk foods such as ultra-processed food, beverages, and snacks were evaluated across the 17 studies. Table 1 illustrates detailed characteristics of records including the age of participants and provenance of studies. All of the included studies have good quality.

Table 1 Characteristics of the included studies
Full size table

Qualitative synthesis

Most of the included studies concordantly showed at least a single significant link between junk food consumption and psychological outcomes. This was despite their use of different measures of association, dissimilar exposure duration and outcomes, and heterogenous definitions, all of which made it challenging to draw conclusions from the qualitative synthesis (summarized in Table 2). Nevertheless, findings from some studies were discordant. For instance, while Sangsefidi et al. (2020) and Chaplin et al. (2011) demonstrated a significant association between stress and snack intake, Almajwal et al. (2016) and Zenk et al. (2014) reported non-significant findings, despite the use of similar measures of association and comparable adjustments for covariates. Although a notable number of studies showed a significant link between junk food intake and psychological disorders, the level of disagreement across studies meant that a meta-analysis was essential in order to clarify this relationship.

Table 2 Findings of qualitative synthesis
Full size table

Quantitative synthesis

Pooling OR in cross-sectional studies

Four cross-sectional studies (n = 13,500) demonstrated that junk food consumption was associated with increased stress (pooled OR = 1.33, 95% CI: 1.02 to 1.65). This finding shows a significant association; however, a notable level of heterogeneity was observed (I² = 74.3%, p = 0.009) (Fig. 2; Table 3). Also, six cross-sectional studies, including 74,127 participants, illustrated a significant association between junk food consumption and depression, with a pooled OR of 1.16 (95% CI: 1.04 to 1.28) (Fig. 2). Overall, junk food consumption indicated a significant association with increased odds of mental health problems (OR = 1.15, 95% CI: 1.06 to 1.23). The Egger’s test for small-study effects indicated evidence of publication bias (p > 0.001). To address this bias, a trim and fill analysis was conducted, resulting in an adjusted OR of 1.11 (95% CI: 0.95 to 1.30). Funnel plot is presented in Fig. 3.

Fig. 2
figure 2

Junk food consumption (unhealthy snacks and sweetened beverages) and odds of having depression and stress in cross-sectional studies

Full size image
Fig. 3
figure 3

Funnel plot, using data from cross-sectional studies investigating the association between junk food consumption and mental health problems

Full size image
Table 3 Findings of Quantitative Synthesis
Full size table

Pooling PR in cross-sectional studies

Two cross-sectional studies focusing on stress with a combined sample size of 2,232 participants reported a PR of 1.31 (95% CI: 1.07–1.55) (Fig. 4).

Fig. 4
figure 4

Association between junk foods consumption and having stress in cross-sectional studies

Full size image

Pooling OR in cohort studies

Pooling results of cohort studies showed that junk food consumption significantly increases the odds of depression by 15% (OR = 1.15; 95% CI: 1.06 to 1.24). After inclusion of the single cohort study that considered stress as its outcome, the overall OR of junk foods consumption and mental disorders was 1.16 (OR = 1.16, 95% CI: 1.07 to 1.24) (Fig. 5).

Fig. 5
figure 5

Association between junk foods consumption and having mental health problems in cohort studies

Full size image

Although Egger’s test for small-study effects yielded a bias coefficient of 2.53, standard error of 1.19, and a p-value of 0.07, trim and fill analysis did not impute any studies, and the overall OR remained unchanged. Figure 6 demonstrates the funnel plot.

Fig. 6
figure 6

Funnel plot, using data from cohort studies investigating the association between junk food consumption and mental health problems

Full size image

Pooling HR in cohort studies

Aggregating two cohort studies with 41,637 participants showed an HR of 1.30 (95% CI: 1.15 to 1.45) for depression, demonstrating a significant risk increase (Fig. 7). Remarkably, these studies showed no heterogeneity (I² = 0.0%, p = 0.81) or publication bias.

Fig. 7
figure 7

Junk food consumption and risk of depression in cohort studies

Full size image

Discussion

The meta-analysis reported in the present study showed that high consumption of junk foods was significantly associated with increased risks of depression. In addition, higher junk food consumption was associated with increased odds of depression and psychological stress. This association between consumption of food with low nutritional value and mental health was demonstrated in multiple studies on different populations and cultures [17,18,19].

Meta-analysis of prospective studies showed that increased risk of subsequent depression and adverse mental health outcomes were correlated with higher ultra-processed food intake [20]. According to meta analysis incorporating seven studies, junk food consumption increased the risk of experiencing mental illness symptoms [21]. For example, one study reporting outcomes for 1591 adults, demonstrated that high consumption of fast foods and processed foods was associated with anxiety, nervousness, restlessness, lack of motivation and depressive symptoms [22]. In another study, weight gain due to unhealthy eating was associated with deterioration in mental health in 404 adults during the second year of the COVID-19 pandemic [23]. Our findings are consistent with a recent systematic review and dose-response meta-analysis that included 26 studies and 260,385 participants from twelve countries, which showed that ultra-processed food consumption increased risk of depression [24].

Epidemiological data suggests that unhealthy food consumption may be associated with poorer mental health through its adverse effects on inflammatory processes, nutritional status, and neurotransmitter function. Inflammation has previously been associated with underlying biological bases for depression [25]. Several observational and meta-analysis studies have demonstrated an inverse association between the consumption of healthy foods including vegetables, fruits, whole-grain and fish, with depressive symptoms [26,27,28,29]. Healthy dietary patterns include a significant amount of tryptophan, an essential amino acid and precursor to serotonin; evidence shows that reduction in the availability of serotonin is associated with depression [30, 31].

The adoption of western dietary patterns that regularly include junk foods and fast foods can increase the probability of developing inflammatory and cardiovascular diseases. Inflammatory conditions are related to mental health disorders including depression, stress and anxiety [32, 33]. In addition, life stressors may augment the interconnection between depressive mood and unhealthy dietary patterns through activation of the brain’s reward system by foods that are high in sugar, fat, and salt [34].

There is also evidence that brain-derived neurotrophic factor (BDNF) may be reduced by consumption of a high fat diet. BDNF is associated with supporting existing neurons and the production of new neurons and implicated in the pathogenesis of depressive disorder. A reduction in BDNF impairs synaptic and cognitive function and neuronal growth, contributing to the development of psychological disorders [35]. Western-type diets include a higher amount of polyunsaturated omega-6 fatty acids, which increase proinflammatory eicosanoids, and decrease BDNF and neuronal membrane fluidity [36]. This suggests that the adverse effects of junk and fast foods on mental health might be associated with the high content of unhealthy fats contained in these foods [4]. Moreover, intake of high amounts of sugar through consumption of sweet drinks and snacks can lead to endothelial dysfunction, inflammation, and exaggerated insulin production that may also influence mood [37,38,39,40].

Mood disorder may itself influence diet, with some studies reporting that patients with depression consume a large amount of carbohydrate-fat-rich foods during their depressive episodes [41,42,43]. Serotonin, an important neurotransmitter for regulating mood, may play a prominent role in this respect given that the sole source of its precursor, tryptophan, is through the diet [44].

The consumption of ultra-processed foods is positively correlated with unhealthy eating habits, including lower intake of fruits and vegetables and higher intake of sweet foods or beverages [8, 45]. It is notable that ultra-processed foods contain additives as well as molecules that are generated by high-temperature heating. These can alter gut microbiota composition and reduce nutrient absorption [46]. Some studies have explored the association between the gut microbiome and mental health [47,48,49], with animal studies suggesting that food additives might increase symptoms of and susceptibility to anxiety and depression via changes of gut microbiota composition [50, 51].

The present paper found that the outcomes of studies selected for the meta-analysis were not always in agreement. This may have been due to confounding factors such as past history of depression or negative life events not being included in the analysis, differences in study designs, sample sizes or population characteristics, non-homogeneous assessment of dietary patterns, and inconsistencies in the evaluation of psychological disorders including the use of different diagnostic criteria to define mental health status.

On the other side, some studies have reported that mental health disorders including depression and psychological stress may reduce an individual’s motivation to eat healthy foods and sometime lead to overeating [17], skipping main meals and replacing them with high calories foods [30]. Some individuals consume high energy and fatty foods during stressful situations, choosing these more palatable foods as an unconscious or deliberate strategy to change their energy levels and mood [52, 53]. Stress affects neuroendocrine function by activating the hypothalamic-pituitary-adrenal (HPA) axis, increasing the secretion of glucocorticoids. These change glucose metabolism, promote insulin resistance, and alter the secretion of appetite-related hormones. All of these factors contribute to the propensity to eat more high-calorie palatable food [12]. However, there are also studies that report no differences in eating patterns under stressful and non-stressful conditions [54, 55]. The analysis presented in the present study cannot be used to demonstrate causality. On the basis of the evidence, it is plausible that there is a bidirectional relationship between junk food consumption and mental health [17]. It remains unclear whether the quality of food choices affects susceptibility to poorer mental health outcomes, and/or the experience of unpleasant emotions influences the quality of food selection [30]. Evidence for a causal pathway is unclear and needs to be further investigated in well-controlled longitudinal studies. Our meta-analysis on cross-sectional studies showed an association between junk food consumption and increased odds of having stress and depression. Besides, meta-analysis on cohort studies demonstrated that junk foods consumption increases the risk of developing stress and depression.

Strengths and limitations

As the main strength of our study, we have comprehensively and specifically evaluated earlier findings regarding the association between junk food consumption and mental health status in adults. The present study has some limitations arising from the studies selected for meta-analysis. Inconsistencies in design of studies such as the ways that diet is assessed using different dietary questionnaire tools, the influence of seasonal and hormonal variations of depressive symptoms, and the use of different diagnostic criteria for defining mental health status is one of the limitations of this study. Despite the association shown between consumption of junk foods and mental health disorders, the strength of the associations and number of documents included in this study is unable to demonstrate causality.

Conclusion

The present study supports the conclusion that consumption of junk foods that are high in fat and sugar content and of low nutritive value are associated with poorer mental health in adults. Further studies utilizing a longitudinal design are needed to better determine the directionality and effect size of junk food consumption on psychological disorders. Moreover, more studies are warranted to assess the mechanisms involved in this relationship to provide scientific support for changes in public health policies.