Depression increases the risk of asthma: Evidence from a cross-sectional study and two-sample Mendelian randomization analysis

Abstract

Depression and asthma are both serious public health issues and are prone to comorbidity. However, the correlation and causality between these conditions remain unclear. Data from the 2005 to 2023 National Health and Nutrition Examination Survey were analyzed. Weighted multivariate logistic regression was used to assess the depression–asthma association, including both crude (unadjusted) and adjusted models (controlling for age, sex, and other covariates). Subgroup analyses were performed to explore risk variations. Mendelian randomization (MR) analyses were primarily conducted using genetic instruments derived from European populations, with the random effect inverse-variance weighted method as the main approach, supplemented by weighted median, MR-Egger, simple mode, and weighted mode analyses. Sensitivity analyses examined heterogeneity, horizontal pleiotropy, and result robustness. The observational study included 38,127 participants. Moderate (OR = 1.56, 95% CI 1.35–1.80, P < .001), moderately severe (OR = 1.97, 95% CI 1.59–2.44, P < .001), and severe (OR = 2.02, 95% CI 1.48–2.76, P < .001) depression were associated with increased asthma risk (P for trend < .001). Interaction effects were noted by age and gender. MR analysis indicated genetically predicted depression causally increases asthma risk (OR = 1.29, 95% CI 1.19–1.40, P < .001) with no significant horizontal pleiotropy in two-sample MR studies. Depression may be a causal factor for asthma. Increased attention is needed for asthma risk in depressed patients, especially men and the elderly. Further validation in large, prospective cohort studies is required.

1. Introduction

Asthma is a chronic respiratory condition characterized by inflammation and narrowing of the airways, leading to symptoms such as wheezing, shortness of breath, chest tightness, and coughing. It is a major public health issue worldwide. The global prevalence of asthma was estimated to be 9.8% in 2019,^[1] and it was the second leading cause of death among chronic respiratory diseases, estimated to affect 5 to 10% of people worldwide.^[2] According to the Global Burden of Disease Study, asthma impacts approximately 300 million people globally, with prevalence rates varying by age and sex.^[3] In the United States, asthma has a particularly significant impact on the population, affecting over 24 million people and leading to over 3517 deaths and 0.9 million emergency room visits in 2021. The burden of asthma extends beyond individual health, encompassing substantial economic costs related to healthcare utilization, medication, and lost productivity. In addition to these direct costs, asthma exacerbates the overall health burden by contributing to impaired quality of life and increased risk of comorbid conditions. Effective management and prevention strategies are therefore crucial for mitigating these impacts.

There is evidence that asthma is a representative psychosomatic disorder of the respiratory system. Recent research has increasingly focused on the complex interactions between asthma and mental health, particularly depression. The combined effects of psychological and social factors with allergic antigens may be a significant trigger for asthma attacks. Among these, psychological stress is a crucial factor in the onset of bronchial asthma attacks. Accurately understanding the relationship between depression and asthma may help improve the rate of complete asthma control and identify new potential treatments for patients with these conditions. Research has found that the prevalence of depression is higher among asthma patients compared to the general population.^[4] Additionally, studies have shown that anxiety and depression can increase the incidence and mortality of asthma, and are significant risk factors for exacerbation of asthma.^[5,6] A meta-analysis has indicated that comorbid depression significantly increases hospitalization rates and unplanned visits for asthma patients due to poor asthma control.^[7] However, existing studies have some limitations. Firstly, many studies have small sample sizes and lack population representativeness, which may affect the generalizability of the results. Secondly, most research focuses on the impact of asthma on depression, with relatively few studies exploring the reverse impact of depression on asthma. Further large-scale studies are needed to confirm the relationship between the 2 diseases.

The National Health and Nutrition Examination Survey (NHANES) is a comprehensive program designed to assess the health and nutritional status of adults and children in the United States. Conducted by the National Center for Health Statistics, NHANES combines interviews, physical examinations, and laboratory tests to gather data on a wide range of health topics. The cross-sectional survey is notable for its representative sampling, ensuring that its findings are generalizable to the entire American population. Mendelian randomization (MR) analysis, which uses genetic variants as instrumental variables (IVs), explores the association between exposure and outcome based on genetic epidemiology.^[8] Because exposed IVs are randomly assigned during conception and are not expected to be influenced by disease state, MR studies may assess the causality by eliminating unobserved confounders and reverse causality. Furthermore, shortcomings of randomized controlled trials, such as high cost and prolonged duration, can be compensated in MR analysis.

Existing evidence suggests potential bidirectional associations between depression and asthma, possibly mediated by shared biological pathways. This study analyzed the association between depression and asthma according to cross-sectional data from the NHANES database. Apart from this, MR analysis was further performed to assess the genetic causal relationship between the 2 diseases, aiming to provide epidemiological evidence supporting potential new preventive and control strategies for asthma.

2. Materials and methods

2.1. Cross‑sectional study design

2.1.1. Data sources and participants

Data for this cross-sectional study were derived from NHANES 2005 to 2023 cycles, and mainly included 3 sections: demographic, examination, and questionnaire. An aggregate of 88,429 participants initially. The exclusion criteria encompassed the following: individuals lacking the Patient Health Questionnaire-9 (PHQ-9), individuals lacking asthma diagnosis information, individuals with missing covariates. From the initial 88,429 participants, the current study excluded 43,238 lacking PHQ-9 depression questionnaire data and 6 additional participants missing asthma diagnosis information, leaving 45,185 eligible individuals. Among these, this study further excluded 7058 with missing values for any covariates included in the adjusted models. Overall, 38,127 individuals met the criteria for analysis. All participants in the survey provided written informed consent at recruitment, hence no additional ethical approval was required as the data for our current study were obtained from the publicly available database.

2.1.2. Definition of depression and asthma

Depression were assessed by the PHQ-9, a screening tool that consists of 9 symptomatic questions ranging from 0 (not at all) to 3 (nearly every day) on the frequency of depressive symptoms over the past 2 weeks, resulting a total score of 0 to 27 points. In line with prior research, we utilized 10 as the cutoff value for clinically significant depression, with 10 to 14 as moderate depression, 15 to 19 as moderately severe depression, and 20 to 27 as severe depression.^[9] The total score was also be analyzed as a continuous variable to further validate our findings. Asthma diagnosis relies on 2 specific questions from the Medical Condition questionnaire: ever been told you have asthma? Had asthma attack in past year? If participants answered YES to either of the above questions, they are considered to have asthma. Use antiasthmatic drug.

2.2. MR study design

2.2.1. Data sources and participants

This study utilized a two-sample MR approach to investigate the causal relationship between depression and asthma. In MR studies, genetic variation must satisfy 3 core assumptions: be strongly associated with the exposure of interest (relevance assumption). Not be associated with confounders of the exposure-outcome (independence assumption). Influence the outcome only through the exposure (exclusion restriction assumption). The depression data were acquired from the Psychiatric Genomic Consortium major depression genome-wide association studies (GWAS), with the GWAS ID being ieu-b-102.^[10] Participants included 500,199 European individuals consisting of 170,756 cases and 329,443 controls. The asthma data were sourced from the IEU Open GWAS database, with the GWAS ID being ebi-a-GCST90014325,^[11] which comprises European samples consisting of 56,167 cases and 352,255 controls. This MR study was conducted using a previously published, publicly available GWAS abstract dataset. All participants provided informed consent in the corresponding GWAS.

2.2.2. Selection of IVs

IVs that met the genome-wide significance threshold (P < 5 × 10⁻⁸) were extracted from the GWAS summary statistics of depression. To avoid the effect of linkage disequilibrium, a strict clumping threshold was adopted: linkage disequilibrium parameter r² < 0.001, genetic distance kb > 10,000. The palindromic single nucleotide polymorphisms (SNPs), as well as the SNPs related to confounding factors and outcomes, were excluded from this study. In addition, F statistic values were calculated to assess the strength of each IV using the formulation: F = (n − 2) × R²/(1 − R²) as reported, where R² was calculated to assess the correlation with exposure and n is the sample size. To calculate R², the following formula was employed: R²=β² × (1 − EAF) × 2EAF, where EAF is the effect allele frequency, β is the estimated genetic effect on exposure. To limit bias from weak IVs bias, an F statistic value exceeding 10 has been recognized as a robust IV and recommended to be used in MR study.

2.3. Statistical analysis

In the observational study deriving data from NHANES, all analyses were performed using weighted data to guarantee estimates that are nationally representative. Continuous variables were expressed as means along with standard error and are compared using analysis of variance for variables that follow a normal distribution, otherwise, non-normally distributed continuous variables were presented as medians with interquartile range and compared using the Kruskal–Wallis test. Categorical variables were described using frequencies (n) and percentages (%) and compared using Chi-square tests. Multivariate logistic regression analysis was employed to evaluate the relationship between depression and the risk of asthma. The median of each depression category was used as a continuous variable to compute P for trends. Subgroup analyses were performed for covariates such as age, gender, and smoking.

For the causal analysis between exposure and outcome, the random effect inverse variance weighting (IVW) was used as the primary analysis method to reveal a causal relationship between depression and asthma, supplemented by MR-Egger, weighted median, simple mode, and weighted mode. Heterogeneity and directional pleiotropy were assessed using Cochran Q test and Egger-intercept test, respectively. Sensitivity analysis was conducted by the leave-one-out approach. A forest plot was used to show the findings of single and multiple SNP analyses, and the funnel plot was evaluated for assessing possible directional pleiotropy. Statistical analysis was performed using the R software (version 4.4.2; R Foundation for Statistical Computing, Vienna, Austria). All the statistical tests performed as two-sided, with P-value < .05 was considered statistically significant.

3. Results

3.1. Cross-sectional study

A total of 38,127 NHANES participants were included in this study. PHQ-9 scores were higher in asthma patients than in healthy controls (Table 1). Compared to healthy participants, the proportion of asthma patients with comorbid depression is higher. Moreover, there were statistically significant differences between asthma patients and healthy controls in age, sex, race, income, BMI, education level, and smoking status (P < .05).

Table 1.

Baseline characteristics of participants by asthma status.

Characteristics	Total (n = 38127)	No asthma (n = 32306)	Asthma (n = 5821)	P-value
Age (yr)	47.53 (0.23)	47.95 (0.24)	45.21 (0.33)	<.001
Gender (n, %)				<.001
Female	19,614 (51.44)	16,197 (50.14)	3417 (58.70)
Male	18,513 (48.56)	16,109 (49.86)	2404 (41.30)
Race (n, %)				<.001
Mexican American	5175 (13.57)	4698 (14.54)	477 (8.19)
Non-Hispanic Black	7859 (20.61)	6481 (20.06)	1378 (23.67)
Non-Hispanic White	17,524 (45.96)	14,703 (45.51)	2821 (48.46)
Other Hispanic	3513 (9.21)	2940 (9.10)	573 (9.84)
Other Race	4056 (10.65)	3484 (10.79)	572 (9.84)
Education (n, %)				.04
Below high school	8071 (21.17)	6967 (21.57)	1104 (18.97)
High school graduate	8685 (22.78)	7390 (22.87)	1295 (22.24)
Above high school	21,371 (56.05)	17,949 (55.56)	3422 (58.79)
PIR (n, %)				<.001
<1	7649 (20.06)	6242 (19.32)	1407 (24.17)
≥1	30,478 (79.94)	26,064 (80.68)	4414 (75.83)
BMI (kg/m2)	29.24 (0.07)	29.00 (0.07)	30.56 (0.16)	<.001
Smoke (n, %)				<.001
No	21,064 (55.25)	18,112 (56.06)	2952 (50.71)
Yes	17,063 (44.75)	14,194 (43.94)	2869 (49.29)
PHQ-9 score	2.00 (0.00, 4.00)	2.00 (0.00, 4.00)	3.00 (1.00, 6.00)	<.001
Depression (n, %)				<.001
No	34,589 (90.72)	29,662 (91.82)	4927 (84.64)
Moderate	2228 (5.85)	1700 (5.26)	528 (9.07)
Moderately severe	928 (2.43)	675 (2.09)	253 (4.35)
Severe	382 (1.00)	269 (0.83)	113 (1.94)

BMI = body mass index, PHQ-9 = Patient Health Questionnaire-9, PIR = poverty income ratio.

Table 2 presents the results of weighted multivariate logistic regression in 3 models. PHQ-9 score was positively correlated with the risk of asthma, both in the crude and adjusted models (fully adjusted model: OR = 1.06, 95% CI 1.05–1.06, P < .001). Depression was categorized into 4 levels, and weighted multivariate logistic regression found that compared to healthy participants, those with depression have a higher risk of developing asthma. Furthermore, the severity of depression was positively correlated with the risk of asthma, meaning that the higher the severity, the greater the risk of developing asthma (P for trend < .001).

Table 2.

The weighted odds ratio for the relationship between depression and asthma.

	Crude OR (95% CI)	P-value	Model 1 OR (95% CI)	P-value	Model 2 OR (95% CI)	P-value
Depression
PHQ-9 score	1.07 (1.06, 1.08)	<.001	1.06 (1.06, 1.07)	<.001	1.06 (1.05, 1.06)	<.001
Categories
No	Reference		Reference		Reference
Moderate	1.82 (1.58, 2.09)	<.001	1.71 (1.49, 1.97)	<.001	1.56 (1.35, 1.80)	<.001
Moderately severe	2.32 (1.91, 2.82)	<.001	2.21 (1.80, 2.71)	<.001	1.97 (1.59, 2.44)	<.001
Severe	2.37 (1.79, 3.14)	<.001	2.28 (1.72, 3.02)	<.001	2.02 (1.48, 2.76)	<.001
P for trend		<.001		<.001		<.001

Crude: no covariates were adjusted.

Model 1: age, gender, and race were adjusted.

Model 2: age, gender, race, PIR, BMI, education, and smoke were adjusted.

CI = confidence interval, OR = odds ratio, PHQ-9 = Patient Health Questionnaire-9.

Table 3 displays the results of subgroup analyses examining the relationship between depression and asthma across various populations. This study found that the risk of asthma varies among different populations, specifically: men (OR = 1.069, 95% CI 1.055–1.082, P < .001) and individuals over 65 years old (OR = 1.069, 95% CI 1.048–1.091, P < .001) have a higher risk of comorbid depression and asthma compared to women and adults under 65 years old.

Table 3.

Subgroup analysis for depression and risk of asthma.

Character	OR (95% CI)	P-value	P for interaction
Gender			.026
Female	1.046 (1.036, 1.057)	<.001
Male	1.069 (1.055, 1.082)	<.001
Age (yr)			.032
20–65	1.054 (1.046, 1.063)	<.001
>65	1.069 (1.048, 1.091)	<.001
Smoke			.052
No	1.050 (1.036, 1.063)	<.001
Yes	1.059 (1.048, 1.071)	<.001

CI = confidence interval, OR = odds ratio.

3.2. MR study

After selection, a total of 47 SNPs were included in the follow-up MR analysis after the exclusion of 2 palindromic SNPs obtaining intermediate allele frequencies, with the F statistics of all SNPs >10. The detailed information on selected SNPs is listed in Table S1, Supplemental Digital Content, https://links.lww.com/MD/Q215. Two-sample MR analysis showed a potential causal effect of depression on asthma (IVW: OR = 1.29, 95% CI 1.19–1.40, P < .001). Similar results were obtained from the weighted median, and weighted mode methods (Table 4). However, the causal relationship was not significant in the evaluation conducted by MR-Egger (OR = 1.49, 95% CI 0.88–2.41, P = .149). Sensitivity analyses revealed moderate heterogeneity in the genetic instruments, with Cochran Q tests yielding statistically significant results for both MR-Egger (Q = 64.28, P = .03, I² = 29.98%) and IVW approaches (Q = 64.71, P = .03, I² = 28.91%). The MR-Egger intercept test showed no evidence of significant horizontal pleiotropy in the analysis between depression and asthma risk (intercept = −0.004, P = .58). The forest plot (Fig. 1) and scatter plot (Fig. 2) both displayed the causal estimates deduced from individual IVs. The leave-one-out analyses (Fig. 3) identified no individual IV largely affects the causal magnitude. The funnel plot for the visualization of heterogeneity is shown in Figure 4.

Table 4.

Estimates of the two-sample MR analyses for causal associations between depression and asthma.

Exposure	Outcome	Method	SNPs	OR (95% CI)	P-value
Depression	Asthma	IVW	47	1.29 (1.19–1.40)	<.001
(ieu-b-102)	(ebi-a-GCST90014325)	MR-Egger	47	1.49 (0.88–2.41)	.149
		Weighted median	47	1.30 (1.16–1.45)	<.001
		Simple mode	47	1.28 (1.00–1.63)	.052
		Weighted mode	47	1.28 (1.01–1.63)	.047

95% CI = 95% confidential interval, IVW = inverse variance weighted, MR = Mendelian randomization, OR = odds ratio, SNP = single nucleotide polymorphism.

Figure 1.

Forest plot shows the contribution of SNPs to the overall effect estimate. SNP = single nucleotide polymorphism.

Figure 2.

Scatter plot shows depression–asthma causal effects.

Figure 3.

Forest plot of leave-one-out sensitivity analysis shows the impact of each SNP on the overall causal estimate. SNP = single nucleotide polymorphism.

Figure 4.

Funnel plot shows no significant heterogeneity among the SNPs. SNP = single nucleotide polymorphism.

4. Discussion

This study, through correlation analysis, found that depression is a risk factor for asthma. The risk of asthma attacks is higher among depressed men and elderly patients over 65. The potential genetic causal relationship between depression and asthma was further confirmed by Mendelian analysis, which strengthens the robustness of the study’s findings.

The observational research indicated that regardless of whether depression scores were treated as continuous or categorical variables, there was an association between depression and an increased risk of asthma. This study also found that the higher the severity of depression, the greater the risk of developing asthma. Existing studies have found that depression and asthma are highly comorbid,^[12] and depression is a risk factor for asthma.^[13,14] A cross-sectional study of 176 children suggested that depression is associated with poor asthma control.^[15] These are consistent with the results obtained in this study. Compared with the above studies, this study further explored the genetic association between the 2 diseases based on large sample population data, and the results are more robust.

It is commonly believed that mental stress is associated with subsequent increased asthma exacerbations. While the precise mechanism linking depression and asthma remains unknown, there are multiple potential hypotheses. First, stress-induced activation of the hypothalamic–pituitary–adrenal axis leads to excessive release of stress hormones, including glucocorticoids. This hormonal imbalance can disrupt normal immune function, promoting pro-inflammatory states in the airways while simultaneously impairing the body’s ability to regulate inflammation effectively.^[16,17] Additionally, chronic exposure to these stress hormones may also reduce the effectiveness of corticosteroid treatments in some asthma patients.^[18] Second, stress activates the autonomic nervous system, particularly through increased cholinergic activity. This results in excessive stimulation of airway smooth muscles and mucus-producing cells, leading to bronchoconstriction and increased mucus secretion (hallmark features of asthma exacerbations).^[19,20] Third, the inflammatory state serves as a common link between the 2 conditions. Patients with depression often exhibit systemic inflammation, as evidenced by elevated levels of inflammatory markers (such as IL-1, IL-4, IL-6, etc) in the blood.^[21] Depression may activate the body’s inflammatory response mechanisms, leading to excessive accumulation of these inflammatory factors. These factors can modulate immune responses and airway smooth muscle function, altering the inflammatory state of the airways and increasing airway reactivity. Consequently, asthma patients may be more prone to asthma attacks when facing stress or emotional fluctuations.

This study shows that males with depression are more likely to develop asthma compared to females, highlighting the need to delve into gender differences and their underlying mechanisms. Research indicates that estrogen in females has a protective effect on the immune system, helping to suppress excessive inflammation, whereas males lack this protective mechanism.^[22] In the context of depression, these gender differences may lead to males being more susceptible to immune dysregulation, which can exacerbate airway inflammation and increase the risk of asthma attacks. Additionally, social and psychological factors cannot be overlooked. In many cultures, men are expected to display strength and independence, and this societal pressure may cause them to internalize and conceal emotional distress, leading to worsened psychological problems and subsequently impacting physical health. This interplay between psychological and physiological factors may contribute to a higher incidence of asthma in this population. The study also found that patients over the age of 65 with depression face a higher risk of developing asthma. The aging population is already subject to various age-related physiological changes. As people age, immune system function declines, and airway responsiveness to stimuli increases, all of which are potential risk factors for asthma.^[23] Additionally, elderly patients with both depression and asthma may experience emotional issues that affect their cognitive function and self-management abilities, leading to poor medication adherence and, consequently, inadequate asthma control.^[24–26] Therefore, it is crucial to be more vigilant about the risk of asthma occurrence or poor asthma control in both elderly and depressed patients.

There were several strengths in the study. First, this study combines observational research with MR analysis, overcoming the influences of confounding factors and reverse causation, thus making the results more reliable. Additionally, both our observational research and MR analysis are based on large-scale data, providing sufficient statistical power to estimate the association between depression and asthma. However, this study does have some limitations. First, the NHANES data relies on self-reported information from participants, which may lead to recall bias and reporting bias. If participants do not accurately report their depressive symptoms or asthma symptoms, this could affect the accuracy of the data and the reliability of the study results. Second, although NHANES data has a large sample size, it may not fully represent the characteristics of all individuals with depression, particularly due to the lack of information on children and adolescents with depression. Finally, our study results are based on European and American populations, which limits the generalizability of our findings to other populations.

5. Conclusions

Depression increases the risk of asthma, and this risk varies across different genders and age groups. This finding needs to be validated through clinical randomized trials. In summary, this study provides new insights for the prevention and treatment of psychosomatic disorders, with asthma serving as a representative condition.

Author contributions

Data curation: Manli Ran, Yu Zhang.

Formal analysis: Manli Ran, Yu Zhang.

Methodology: Wang Zhang, Minjian Wang.

Project administration: Siying Luo.

Software: Wang Zhang, Minjian Wang.

Supervision: Siying Luo.

Writing – original draft: Manli Ran, Yu Zhang.

Writing – review & editing: Siying Luo.

Supplementary Material

medi-104-e44972-s001.docx ^{(37.1KB, docx)}