Systematic Review With Meta-analysis

The Prevalence of Anxiety and Depression in Patients With Irritable Bowel Syndrome

Mohammad Zamani; Shaghayegh Alizadeh-Tabari; Vahid Zamani

Disclosures

Aliment Pharmacol Ther. 2019;50(2):132-143. 

In This Article

Results

Search Results, Study Selection and Characteristics

The initial search in the databases identified a total of 14 926 citations and 5847 remained after removing duplicates. After evaluating the title and/or abstract, 5634 articles were excluded because of failure to meet the inclusion criteria. Full-text of the remaining 213 papers was assessed for eligibility; 73 eligible articles were included in the final analyses. The results of search strategy are shown in a flow diagram according to the PRISMA guideline (Figure 1). The characteristics of the included articles are summarised in Tables S1-S4.

Figure 1.

PRISMA flow diagram

Anxiety Symptoms

The information of studies related to anxiety symptoms have been summarised in Table S1. Analysis of 43 articles[19–61] indicated that the pooled prevalence of anxiety symptoms was 39.1% (95% CI: 32.4-45.8; I 2 = 99.1%) in 16 572 IBS patients (Figure 2). Visual inspection of the funnel plot shows some asymmetry (Figure S1). Egger's test also indicated significant publication bias (P = 0.010). Meta-regression analyses indicated that gender (% male, β = 0.002, P = 0.810) and publication year (β = 0.045, P = 0.070) did not explain the heterogeneity in outcomes.

Figure 2.

Pooled prevalence of anxiety symptoms in patients with the irritable bowel syndrome

Subanalysis exhibited that the pooled OR for anxiety symptoms in IBS patients compared with healthy subjects was 3.11 (95% CI: 2.43-3.98) based on 24 papers, with significant heterogeneity between studies (I 2 = 93.6%, P < 0.001) (Figure S2). Subgroup analyses from five studies showed that the pooled prevalence rates of anxiety symptoms were 43.7% (95% CI: 33.2-54.2), 35.5% (95% CI: 26.9-44.1) and 37% (95% CI: 26.9-47) in patients with IBS-C, IBS-D and IBS-M respectively. Subgroup analyses by gender from five articles on males and eight papers on females showed that the pooled prevalence of anxiety symptoms was 31.5% (95% CI: 15.3-47.7) and 47.8% (95% CI: 36-59.6) in males and females with IBS respectively. The pooled OR for anxiety symptoms in male IBS patients compared with female IBS patients was 0.47 (95% CI: 0.18-1.23) based on five studies, with significant heterogeneity between studies (I 2 = 93.4%, P < 0.001). Regarding IBS diagnostic instruments, 4, 15 and 21 studies used Rome I, Rome II and Rome III, and the subgroup analysis by these criteria showed the pooled prevalence rates of 36.5% (95% CI: 14.5-58.5), 41.3% (95% CI: 31.6-51.1) and 43.5% (95% CI: 31.3-55.7) for anxiety symptoms respectively (P = 0.558). No enough data existed to perform subgroup analysis by other diagnostic methods.

Anxiety Disorders

The information of studies related to anxiety disorders have been summarised in Table S2. The pooled prevalence of anxiety disorders was 23% (95% CI: 17.2-28.8; I 2 = 99.9%) among 375 534 IBS patients based on 20 articles[51,62–80] (Figure 3). Visual inspection of the funnel plot did not indicate any potential publication bias (Figure S3). Egger's test also showed no significant evidence of publication bias among the included studies (P = 0.478). Meta-regression analyses indicated that gender (% male, β −= −0.010, P = 0.157) and publication year (β = 0.001, P = 0.981) did not explain the heterogeneity between studies.

Figure 3.

Pooled prevalence of anxiety disorders in patients with the irritable bowel syndrome

Analysis of nine studies showed that the pooled OR for anxiety disorders in IBS patients compared with healthy subjects was 2.52 (95% CI: 1.99-3.20), with significant heterogeneity between studies (I 2 = 97.3%, P < 0.001) (Figure S4). No enough data existed to perform subgroup analysis by gender and IBS subgroup.

Depressive Symptoms

The information of studies related to depressive symptoms has been summarised in Table S3. Analysis of 47 articles[19–42,44–46,48–61,81–86] indicated that the pooled prevalence of depressive symptoms was 28.8% (95% CI: 23.6-34; I 2 = 99.1%) in 22 842 IBS patients (Figure 4). Visual inspection of the funnel plot indicates some asymmetry (Figure S5). However, Egger's test showed no significant evidence of publication bias among the included studies (P = 0.517). Meta-regression analyses indicated that gender (% male, β=−0.016, P = 0.108) and publication year (β = −0.028, P = 0.265) did not explain the heterogeneity in outcomes.

Figure 4.

Pooled prevalence of depressive symptoms in patients with the irritable bowel syndrome

Sub-analysis exhibited that the pooled OR for depressive symptoms in IBS patients compared with healthy subjects was 3.04 (95% CI: 2.37-3.91) based on 24 papers, with significant heterogeneity between studies (I 2 = 93.8%, P < 0.001) (Figure S6). Subgroup analyses from five studies showed that the pooled prevalence rates of depressive symptoms were 41.4% (95% CI: 20.2-62.8), 37.1% (95% CI: 16.7-57.6) and 36.8% (95% CI: 17.3-56.4) in patients with IBS-C, IBS-D and IBS-M respectively. Subgroup analyses by gender from seven articles on males and nine papers in females showed that the pooled prevalence of depressive symptoms was 25.9% (95% CI: 11.9-39.9) and 35.1% (95% CI: 23-47.3) in males and females with IBS respectively. The pooled OR for anxiety symptoms in male IBS patients compared with female IBS patients was 0.84 (95% CI: 0.64-1.11) based on seven studies, with significant heterogeneity between studies (I 2 = 55.1%, P = 0.037). In relation to IBS diagnostic instruments, 4, 16 and 23 studies used Rome I, Rome II and Rome III, and the subgroup analysis by these criteria showed the pooled prevalence rates of 18.5% (95% CI: 7.5-29.5), 30.9% (95% CI: 23.5-38.3) and 32.2% (95% CI: 23.1-41.3) for depressive symptoms respectively (P = 0.046). No enough data existed to perform subgroup analysis by other diagnostic methods.

Depressive Disorders

The information of studies related to depressive disorders has been summarised in Table S4. Pooling data from 19 articles[62,63,66,67,69–73,75,77–80,87–91] showed a prevalence rate of 23.3% (95% CI: 17.2-29.4; I 2 = 99.9%) in 407,967 IBS patients (Figure 5). Visual inspection of the funnel plot did not indicate any potential publication bias (Figure S7). Egger's test also showed no significant evidence of publication bias among the included studies (P = 0.388). Meta-regression analyses indicated that gender (% male, β = −0.002, P = 0.769) and publication year (β =− 0.017, P = 0.494) did not explain the heterogeneity between studies.

Figure 5.

Pooled prevalence of depressive disorders in patients with the irritable bowel syndrome

Analysis of 10 papers indicated that the pooled OR for depressive disorders in IBS patients compared with healthy subjects was 2.72 (95% CI: 2.45-3.02), with significant heterogeneity between studies (I 2 = 92.8%, P < 0.001) (Figure S8). No enough data existed to perform subgroup analysis by gender and IBS subgroup.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....