Psychiatric Comorbidities Among Adult Patients with Disorders of Gut-Brain Interaction: Prevalence and Relationships to Treatment Outcomes

Abstract

Background:

Little is known about the impact of psychiatric comorbidity on pharmacologic treatment outcomes, including neuromodulators (medications targeting the gut-brain axis), among adult patients with disorders of gut-brain interaction (DGBI). Accordingly, we aimed to examine associations between psychiatric comorbidity and DGBI pharmacologic treatment outcomes.

Methods:

In a retrospective study of consecutively referred new patients (N=410; ages 18–90; 73% female) to a tertiary neurogastroenterology clinic in 2016 with follow-up through 2018, relationships between psychiatric illness (any psychiatric illness, anxiety disorders, depressive disorders) and pharmacologic treatment selection (any medication, neuromodulating medication) and treatment outcomes, respectively, were examined using multivariable logistic regression, adjusting for demographics, gastrointestinal (GI) diagnoses, and pre-existing neuromodulator use.

Key Results:

Anxiety disorders (35%) were the most common psychiatric comorbidity, followed by depressive disorders (29%). Patients with anxiety disorders were more likely to be prescribed a neuromodulator by their gastroenterologist (OR=1.72 [95% CI 1.10–2.75]) yet less likely to respond to neuromodulators (OR=0.43 [0.21–0.90]) or any GI medication (OR=0.24 [0.12–0.50]) in fully adjusted analyses. In contrast, depressive disorders were not associated with neuromodulator prescription or response.

Conclusions and Inferences:

Anxiety disorders are common among patients with DGBI and significantly reduce the likelihood of GI pharmacologic treatment response to any medication prescribed, including neuromodulators.

Graphical Abstract

Patients with disorders of gut-brain interaction (DGBI) and comorbid anxiety disorders are less likely to respond to any medication, including neuromodulators, prescribed for their gastrointestinal symptoms.

Introduction

Disorders of gut-brain interaction (DGBI; also known as functional gastrointestinal (GI) disorders) are a vast clinical group, increasingly understood to lie along a spectrum of sensory and motility disturbance.¹ Of the thirty-three DGBI defined in adults by the ROME IV criteria,² irritable bowel syndrome (IBS) is the most common (global prevalence of approximately 11% in adults)³ and most widely studied.² The DGBI are thought to be mediated by any combination of the following mechanisms: motility disturbance, visceral hypersensitivity, altered mucosal and immune function, altered gut microbiota, and altered central nervous system processing.² Conceptualized as resulting from complex interactions between biological, psychological, and social factors,⁴ DGBI are associated with high rates of comorbid psychiatric illness.^4,5

Estimates of psychiatric comorbidity have varied significantly across the spectrum of DGBI.⁵ Anxiety disorders are the most common psychiatric comorbidity;⁴ in IBS, prevalence estimates range from approximately 30–50%.^4,5 Rates of depressive disorders across the DGBI tend to be lower, ranging from 25–30%.^4,5 Some of this variability may be related to treatment setting, with higher rates of psychiatric comorbidity often observed in tertiary as opposed to primary care or community settings.^4,6 Across the spectrum of DGBI, there is evidence that psychiatric comorbidity – or at least the presence of psychiatric symptoms – perpetuates and potentiates GI symptoms, heightens visceral hypersensitivity, and worsens quality of life.^5–7

The treatment approach for DGBI is multifaceted and focused on symptom reduction rather than cure.² For mild or infrequent symptoms, lifestyle modification (e.g., diet and exercise), reassurance, and education about the disorder can at times be sufficient.² For moderate to severe symptoms, a combination of pharmacotherapy and behavioral health interventions is often required.² Pharmacotherapy for DGBI includes the use of medications directed at specific GI symptoms, gut motility, and in some cases, the gut-brain axis (i.e., neuromodulators).^8,9 Neuromodulators, such as antidepressants, antipsychotics, and antiepileptics, exert their effects on the gut-brain axis by acting on the same serotonergic, noradrenergic, and dopaminergic neurotransmitters and receptors as in the central nervous system.⁸ Meta-analyses have examined the efficacy of neuromodulators in IBS, with a recent analysis finding a collective number needed to treat of 4.5 for antidepressants.¹⁰ Preliminary studies of efficacy in other DGBI are limited, but support the use of tricyclic antidepressants (TCAs) for pain predominant conditions, and evidence is emerging for the use of other agents.⁸ It remains to be elucidated, however, if and how comorbid psychiatric disease might impact GI pharmacologic treatment response.

Accordingly, among adult patients with DGBI presenting to a tertiary care neurogastroenterology clinic, we aimed to examine associations between psychiatric illness and GI pharmacologic response, both across all medications prescribed for GI symptoms and for neuromodulators, specifically.

Materials and Methods

Procedure

Participants included consecutively referred adult patients (N=410; aged 18–90 years; 73% female) who presented for initial neurogastroenterology evaluation to a tertiary care academic medical center from January 1, 2016 to December 31, 2016.¹¹ Electronic medical records from the initial 2016 visit, as well as all subsequent hospital and clinic visits (including mental health visits) through May 31, 2018, were reviewed by four independent coders (H.B.M, C.J.S., A.P.B., A.C.K., see Acknowledgements). Demographic data, presenting complaints, GI diagnoses, comorbid diagnoses, all medications targeting GI symptoms (e.g., prokinetic agents, antidiarrheal agents, neuromodulators) introduced during the study time frame, and medication response were recorded.¹² Presenting complaints and GI diagnoses were categorized for analysis purposes (see Supplementary Table 1), as were GI medications and psychiatric diagnoses (grouped per the Diagnostic and Statistical Manual of Mental Disorders (DSM-5)).¹³ A board-certified psychiatrist (E.N.M.) independently and blindly reviewed each chart for psychiatric diagnoses (defined as being noted in at least one patient encounter or in the patient problem list during the study time frame), medications targeting GI symptoms that were prescribed during the study time frame (note: neuromodulators were defined as medications acting on the brain-gut axis including antidepressants, antipsychotics, antiepileptics, and buspirone), report of specific medication side effects (yes/no), and medication benefit (yes/no/insufficient data) as described in the medical record. Interrater reliability was strong (κ=.85) for medication response and moderate (κ=.73) for psychiatric diagnosis. An independent blinded board-certified psychiatrist (C.M.C.) reviewed all disagreements, which were subsequently resolved by consensus. Institutional Review Board approval was obtained prior to study procedure commencement.

Statistical Analysis

Descriptive statistics were used to record psychiatric comorbidities. For descriptive purposes, demographic and clinical characteristics were compared between groups (by psychiatric comorbidity presence) with Wilcoxon-Mann-Whitney tests and chi-square tests for continuous and categorical data, respectively. Exploratory multivariable logistic regression analyses examined relationships between psychiatric illness (any psychiatric illness, anxiety disorders, depressive disorders) and treatment selection (prescribed any medication for GI symptoms, prescribed a neuromodulator) and treatment outcomes (response to any GI medication, response to neuromodulator, side effects), respectively. Covariates were selected a priori based on potential links to treatment selection/response and included demographic characteristics (race/ethnicity, age, sex), GI diagnostic category, and pre-existing neuromodulator use. Covariates were entered into the model in a stepwise fashion (model 1=demographics; model 2=demographics + GI diagnosis; model 3=demographics + GI diagnosis + neuromodulator use at the time of initial consultation). Significance was set at p<.05 given the exploratory nature of the analyses. Analyses were performed using Stata version 15.2 (StataCorp, College Station, TX).

Results

Frequency and Characteristics of Psychiatric Comorbidities Among Patients with DGBI

Of the 410 patient charts reviewed, comorbid psychiatric illness was identified in 226 cases (55%). Characteristics of patients with and without psychiatric comorbidity are summarized in Table 1. Descriptive analyses showed that patients with psychiatric comorbidity were more likely to be female (79% vs 65%, p=.002), diagnosed with an upper (38% vs 27%, p=.014) or lower (61% vs 41%, p<.001) GI disorder, taking a neuromodulator at the time of their initial GI visit (54% vs 16%, p<.001), and referred to psychiatry by their GI provider (16% vs 0%, p<.001) than patients without psychiatric comorbidity. There were no significant differences in age, race/ethnicity, body mass index, presenting symptomatology, prescription of any medication following the GI visit, or clinic follow-up.

Table 1.

Characteristics of DGBI patients with and without psychiatric comorbidity (N=410)

	Total	Psychiatric Comorbidity (n = 226)	No Psychiatric Comorbidity (n = 184)	Test Statistic	p-value
Demographics and psychosocial characteristics; n (%)
Age at consult, years; median (interquartile range)	48.0 (31.0 to 61.0)	48.0 (30.0 to 60.0)	49.0 (32.5 to 62.5)	z=1.18	.24
Female	298 (72.7)	178 (78.8)	120 (65.2)	x2=9.37	.002
White non-Hispanic/Latinoa	341 (84.8)	192 (86.5)	149 (82.8)	x2=1.06	.30
Raceb				x2=2.25	.52
American Indian or Alaska Native	1 (0.2)	1 (0.5)	0 (0.0)	-	-
Asian	19 (4.7)	8 (3.7)	11 (6.2)	-	-
Black or African American	17 (4.2)	9 (4.1)	8 (4.5)	-	-
Native Hawaiian or Pacific Islander	-	-	-	-	-
White	350 (87.1)	196 (89.9)	154 (87.0)	-	-
Ethnicity—Hispanic/Latinoc	22 (5.5)	11 (5.0)	11 (6.2)	x2=0.27	.61
Medical Characteristics; n (%)
Body Mass Index- kg/m2; median (interquartile range)	24.8 (21.6 to 28.6)	24.5 (21.4 to 28.5)	25.3 (22.4 to 28.7)	z=1.28	.20
Presenting symptoms
Eating/weight-related	58 (14.1)	36 (15.9)	22 (12.0)	x2=1.32	.25
Esophageal	201 (49.0)	107 (47.3)	94 (51.1)	x2=0.57	.45
Stomach	187 (45.6)	107 (47.3)	80 (43.5)	x2=0.61	.43
Lower	253 (61.7)	149 (65.9)	104 (56.5)	x2=3.80	.051
Pre-existing neuromodulator use at time of initial GI visit	152 (37.1)	122 (54.0)	30 (16.3)d	x2=61.72	<.001
Following GI Consultation; n (%)
Grouped GI diagnostic categories
Esophageal sensory-related diagnoses	69 (16.8)	31 (13.7)	38 (20.7)	x2=3.49	.062
Esophageal reflux-related diagnoses	108 (26.3)	59 (26.1)	49 (26.6)	x2=0.01	.91
Upper GI diagnoses	135 (32.7)	86 (38.1)	49 (26.6)	x2=5.99	.014
Lower GI diagnoses	214 (52.2)	138 (61.1)	76 (41.3)	x2=15.87	<.001
Chronic abdominal pain	34 (8.3)	23 (10.2)	11 (6.0)	x2=2.35	.13
Prescribed any medication following consult	320 (78.0)	181 (80.1)	139 (75.5)	x2=1.22	.27
Prescribed neuromodulator following consult	214 (52.2)	133 (58.8)	81 (44.0)	x2=8.94	.003
Follow-up visit (yes/no)	290 (70.7)	166 (73.5)	124 (67.4)	x2=1.80	.18
Referred to psychiatry	37 (9.0)	37 (16.4)	0 (0.0)	n/a	<.001

Note: DGBI = disorder of gut-brain interaction. GI=gastrointestinal. Eating/weight-related presenting symptoms include: low weight, weight loss, poor appetite, food aversion. Esophageal presenting symptoms include: reflux, regurgitation, belching, globus, choking, cough, dysphagia, heartburn, chest pain/pressure, other throat-related symptoms. Stomach presenting symptoms include: nausea, vomiting, postprandial fullness, early satiety, epigastric pain, upper abdominal pain. Lower presenting symptoms include: lower abdominal pain, diarrhea, constipation, bloating/distension, fecal incontinence, rectal urgency, rectal pain, anal itching, flatulence. Esophageal sensory-related diagnoses include: globus, dysphagia, non-cardiac chest pain. Esophageal reflux-related diagnoses include: GERD, rumination, chronic belching. Upper GI disorders include: gastroparesis, functional dyspepsia, cyclic vomiting syndrome, chronic unspecified nausea and vomiting. Lower GI disorders include: IBS, chronic constipation, functional diarrhea, abdominal bloating/distension.

^an=402. Race and ethnicity data missing for n=8.

^bn=387. Race data missing for n=23.

^cn=405. Ethnicity data missing for n=5.

^dNeuromodulators were prescribed for indications other than psychiatric diagnosis.

Of the psychiatric comorbidities identified based on chart review (Table 2), anxiety disorders were the most common (35%), followed by depressive disorders (29%), post-traumatic stress disorder (4%), and eating disorders (4%).

Table 2.

Frequency of chart psychiatric diagnoses among DGBI patients (N=410)

Chart Psychiatric Diagnosisa	n (%)
Any psychiatric diagnosis	226 (55.1)
Anxiety disorder	145 (35.4)
Depressive disorder	118 (28.7)
Post-traumatic stress disorder	18 (4.4)
Eating disorderb	18 (4.4)
Bipolar disorder (I or II)	14 (3.4)
Attention deficit hyperactivity disorder	12 (2.9)
Adjustment disorder	10 (2.4)
Autism spectrum disorder	9 (2.2)
Substance use disorder	9 (2.2)
Obsessive compulsive disorder	5 (1.2)
Somatic symptom disorder	5 (1.2)
Otherc	4 (1.0)

Note: DGBI = disorder of gut-brain interaction.

^aPsychiatric diagnoses were grouped per the Diagnostic and Statistical Manual of Mental Disorders (DSM-5).¹³

^bEating disorder diagnoses were identified based on chart review and do not include Avoidant/Restrictive Food Intake Disorder (ARFID). ARFID diagnoses have been examined separately using DSM-5 criteria applied to clinical information extracted from the medical record.¹²

^cOther includes psychotic disorders, personality disorders, and major neurocognitive disorder with behavioral disturbance, grouped to prevent potential patient identification.

Relationships between Psychiatric Comorbidities and Treatment Selection

Prescribed any medication.

On multivariable logistic regression, no significant relationships were identified between likelihood of being prescribed any medication for GI symptoms following the initial GI visit and any psychiatric diagnosis, anxiety disorders, or depressive disorders, respectively (Table 3).

Table 3.

Multivariable associations between psychiatric illness and treatment outcomes (N=410)

Predictor	Model	OR (95% CI)	SE	p-value
Prescribed any GI medication
Any psychiatric diagnosis	1	1.27 (0.78–2.10)	.31	.33
	2	1.01 (0.61–1.68)	.26	.97
	3	0.83 (0.48–1.44)	.23	.51
Anxiety disorder	1	1.27 (0.75–2.14)	.34	.37
	2	1.06 (0.61–1.82)	.29	.84
	3	0.94 (0.53–1.64)	.27	.82
Depressive disorder	1	1.32 (0.76–2.29)	.37	.33
	2	1.08 (0.61–1.92)	.32	.79
	3	0.90 (0.49–1.65)	.28	.72
Prescribed any neuromodulator
Any psychiatric diagnosis	1	1.68 (1.12–2.54)	.35	.013*
	2	1.38 (0.89–2.14)	.31	.15
	3	1.19 (0.75–1.89)	.28	.46
Anxiety disorder	1	2.01 (1.30–3.11)	.45	.002**
	2	1.88(1.19–2.96)	.44	.007**
	3	1.72 (1.10–2.75)	.41	.023*
Depressive disorder	1	1.92 (1.21–3.05)	.45	.005**
	2	1.63 (1.01–2.64)	.40	.045*
	3	1.42 (0.86–2.36)	.37	.18
Response to any GI medication
Any psychiatric diagnosis	1	0.27 (0.13–0.58)	.11	.001**
	2	0.29 (0.14–0.64)	.12	.002**
	3	0.24 (0.11–0.56)	.10	.001**
Anxiety disorder	1	0.27 (0.14–0.53)	.92	<.001**
	2	0.28 (0.14–0.55)	.97	<.001**
	3	0.24 (0.12–0.50)	.89	<.001**
Depressive disorder	1	0.61 (0.31–1.17)	.20	.14
	2	0.70 (0.35–1.40)	.25	.32
	3	0.66 (0.32–1.38)	.25	.27
Response to any neuromodulator
Any psychiatric diagnosis	1	0.49 (0.24–1.01)	.18	.055†
	2	0.53 (0.25–1.11)	.20	.092†
	3	0.51 (0.24–1.10)	.20	.087†
Anxiety disorder	1	0.44 (0.22–0.88)	.15	.019*
	2	0.45 (0.22–0.91)	.16	.027*
	3	0.43 (0.21–0.90)	.16	.025*
Depressive disorder	1	0.76 (0.38–1.54)	.27	.45
	2	0.91 (0.43–1.93)	.35	.81
	3	0.92 (0.42–2.00)	.36	.83
Reported side effects
Any psychiatric diagnosis	1	1.49 (0.89–2.51)	.40	.13
	2	1.31 (0.76–2.24)	.36	.33
	3	1.57 (0.89–2.80)	.46	.12
Anxiety disorder	1	1.41 (0.84–2.36)	.37	.19
	2	1.33 (0.79–2.26)	.36	.29
	3	1.52 (0.88–2.64)	.43	.14
Depressive disorder	1	0.95 (0.55–1.63)	.26	.84
	2	0.80 (0.45–1.42)	.23	.45
	3	0.92 (0.50–1.70)	.29	.80

Note: GI=gastrointestinal.

Model 1: Adjusting for demographic characteristics (race/ethnicity, age, sex).

Model 2: Adjusting for model 1 predictors/covariates plus five grouped GI diagnostic categories including esophageal sensory-related diagnoses (globus, dysphagia, non-cardiac chest pain), esophageal reflux-related diagnoses (gastroesophageal reflux disease, rumination, chronic belching), upper GI disorders (gastroparesis, functional dyspepsia, cyclic vomiting syndrome, chronic unspecified nausea and vomiting), lower GI disorders (irritable bowel syndrome, chronic constipation, functional diarrhea, abdominal bloating/distension) and chronic abdominal pain, and pre-existing neuromodulator use.

Model 3: Adjusting for model 2 predictors/covariates plus pre-existing neuromodulator prescription.

^**p<.01

^*p<.05

^†p<.1

Prescribed a neuromodulator.

Univariate analyses showed that patients with psychiatric comorbidity were more likely to be prescribed a neuromodulator following their GI visit (59% vs 44%, p=.003; see Table 1). Multivariable analysis showed that patients with any psychiatric diagnosis (OR 1.68 [1.12–2.54], p=.013), anxiety disorders (OR 2.01 [1.30–3.11], p=.002), and depressive disorders (OR 1.92 [1.21–3.05], p=.005) were all more likely to be prescribed a neuromodulator when adjusting for demographic characteristics only (race/ethnicity, age, sex). However, in fully adjusted analyses, accounting for demographic characteristics, GI diagnoses, and pre-existing neuromodulator prescription, only anxiety disorders were significantly associated with new neuromodulator prescription (OR 1.72 [1.10–2.75], p=.023).

Relationships between Psychiatric Comorbidities and Treatment Outcomes

Response to any medication.

Multivariable logistic regression showed that patients with any psychiatric diagnosis (OR 0.24 [0.11–0.56], p=.001) were significantly less likely to respond to any medication, even when accounting for demographic characteristics, GI diagnoses, and pre-existing neuromodulator prescription. Lower likelihood of treatment response was also seen for anxiety disorders (OR 0.24 [0.12–0.50], p<.001) specifically, but not depressive disorders.

Response to any neuromodulator.

Patients with any psychiatric diagnosis trended toward being less likely to respond to any neuromodulator, though this was not statistically significant (OR 0.51 [0.24–1.10], p=.087). In fully adjusted analyses, lower likelihood of neuromodulator response was seen for anxiety disorders (OR 0.43 [0.21–0.90], p=.025), but not depressive disorders.

Reported side effects.

There were no significant relationships between report of side effects and any psychiatric diagnosis, anxiety disorders, or depressive disorders, respectively.

Discussion

Despite high rates of psychiatric comorbidity across the DGBI spectrum,⁵ the impact of psychiatric illness on GI pharmacologic treatment response has not previously been described. We sought to evaluate the pharmacologic impact of psychiatric illness among adult patients presenting for neurogastroenterology evaluation. Comorbid psychiatric illness was identified in over half (55%) of all patients, with anxiety disorders being the most common (35%). Patients with comorbid anxiety disorders were more likely to be prescribed a neuromodulator, yet less likely to respond to either neuromodulators or any medication prescribed for their GI symptoms during the nearly 2.5-year study time frame. In contrast, no significant relationships were found between depressive disorders and pharmacologic treatment selection or response when adjusting for demographic characteristics, GI diagnoses, and pre-existing neuromodulator prescription. To our knowledge, this represents one of the first examinations of the impact of comorbid psychiatric illness on GI pharmacologic treatment response among adult patients with DGBI.

Our finding that anxiety disorders were the most common (35%) comorbid psychiatric diagnosis across the DGBI spectrum extends the existing literature.^5,14 Anxiety disorders are the most common psychiatric comorbidity in IBS, with rates of approximately 30–50% across treatment settings.^4,5 We identified similarly high rates in our study’s broad population including disorders across the DGBI spectrum.⁵ Interestingly, despite the high rates of psychiatric comorbidity in this population, relatively few referrals were made to outpatient psychiatry (see Table 1), which may reflect involvement of existing mental health treaters, lack of perceived need, long wait times, or other barriers to referral that should be identified in future investigations.

That comorbid anxiety disorders – but not depressive disorders – were associated with worse GI pharmacologic treatment response in our analyses may be reflective of the unique mechanistic role of anxiety in DGBI. It has long been postulated that vulnerability to anxiety disorders and DGBI may share similar pathways,⁴ and a recent genome-wide analysis identified shared genetic susceptibilities.¹⁵ Comorbid anxiety disorders, over all other psychiatric disorders, are associated with the greatest GI symptom burden among DGBI patients,¹⁶ and may potentiate GI-specific anxiety, a key mediator of behavioral treatment outcomes in DGBI such as IBS.¹⁷ Though both depression and anxiety have been shown to contribute to heightened visceral hypersensitivity in IBS,⁷ only comorbid anxiety disorders have been associated with impaired gastric accommodation,¹⁸ increased gastric sensitivity,¹⁹ and modulation of visceral afferent input²⁰ in functional dyspepsia, for example. Though one might have imagined that patients with comorbid anxiety disorders would respond more robustly to neuromodulation (if the neuromodulating agent has both GI and psychiatric efficacy), our findings suggest otherwise, and highlight the likely complex relationship between anxiety and GI function, which current medications may not adequately target.

Study limitations include the retrospective design and use of chart history to identify comorbid psychiatric diagnoses and treatment response. Assessment for psychiatric comorbidity, however, is part of standard neurogastroenterology care in our center and two blinded psychiatrists independently coded psychiatric diagnoses. Additionally, even in this tertiary care population that may have potentially been enriched for psychiatric comorbidity, the prevalence of comorbid psychiatric disorders was consistent with prior research (particularly when considering the entirety of the DGBI spectrum).⁵ Some treatment response granularity may have been lost with our binary pharmacologic treatment response outcome (i.e., any improvement was defined as improvement), but this was necessary for accuracy, as degree of improvement was not consistently described in the medical record. Lastly, there may be other factors associated with pharmacologic treatment response (e.g., duration or severity of GI or psychiatric symptoms, neuromodulator dose, presence of comorbid pain disorders, participation in other therapies) that were not captured in this retrospective study, and which should be explored in future investigations.

Our findings enhance our understanding of the relationship between psychiatric illness and the DGBI spectrum, providing novel information about GI pharmacologic treatment selection and response. As the presence of comorbid anxiety disorders appears to reduce GI pharmacologic treatment response, screening for psychiatric illness, particularly for anxiety disorders, may be key. It may be that targeted treatment of comorbid anxiety disorders through increased referral to mental health professionals and/or the use of integrative, team-based approaches incorporating GI psychologists/psychiatrists and non-pharmacologic treatment modalities (e.g., cognitive-behavioral therapy) could have the potential to lead to better GI outcomes. Further research is needed to clarify the mechanism underlying the relationship between comorbid anxiety disorders and GI pharmacologic treatment response, and to determine which approaches may be most effective at improving GI outcomes in the face of comorbid anxiety.

Key Points.

1. Anxiety disorders were the most common (35%) psychiatric comorbidity in a sample of adult patients with disorders of gut-brain interaction, consistent with prior research.

2. The presence of anxiety disorders significantly reduced the likelihood of gastrointestinal pharmacologic treatment response to any medication prescribed for gastrointestinal symptoms, including neuromodulators, in patients with disorders of gut-brain interaction.

3. Gastrointestinal pharmacologic treatment response in patients with disorders of gut-brain interaction may benefit from early identification and targeted treatment of comorbid psychiatric illness, particularly anxiety disorders. Further research is needed to inform these recommendations.