Mental Health, Work Presenteeism, and Exercise in Inflammatory Bowel Disease

Abstract

Background

Chronic diseases, such as IBD, can lead to anxiety and depression which can have a significant impact on productivity at work [presenteeism]. The aim of this study was to assess the prevalence of depression/anxiety, presenteeism and exercise levels among IBD patients.

Methods

This was a multicentre study whereby adult IBD patients, in clinical remission, were asked to answer a questionnaire anonymously. Hospital Anxiety and Depression Score [HADS], Stanford Presenteeism Scale [SPS-6] and Godin Exercise Score were also collected.

Results

A total of 585 patients were recruited. The majority had Crohn’s disease [CD, 62.2%] and were male [53.0%], with a median age of 39 years [IQR 30-49]. A psychiatric diagnosis was present in 10.8% of patients prior to their IBD diagnosis. A further 14.2% of patients were psychiatrically diagnosed after IBD diagnosis, this being commoner in CD patients [41.6% of CD, p <0.01]. A raised HADS-Anxiety or a HADS-Depression score ≥8 was present in 46.1% of patients, with 27.4% having a score ≥11. Low presenteeism at work was present in 34.0%. Patients diagnosed with depression/anxiety had a more sedentary lifestyle [p <0.01], lower presenteeism at work [p <0.01] and a higher rate of unemployment [p <0.01].

Conclusions

A significant percentage of IBD patients in remission suffer from anxiety and/or depression. Risk factors for these are CD, female gender, use of biologic medications, long-standing and/or perianal disease. Depression/anxiety was associated with a sedentary lifestyle, lower presenteeism at work and unemployment. Validated screening tools and appropriate referrals to psychologists and/or psychiatrists should be employed within IBD clinics.

1. Introduction

Inflammatory bowel diseases [IBD], including Crohn’s disease [CD] and ulcerative colitis [UC], are chronic relapsing conditions causing inflammation of the gastrointestinal tract and may have a substantial effect on an individual’s quality of life.^1–3 Despite improvements in treatment, 20-25% of patients experience ongoing symptoms.⁴

Anxiety and depression rates are higher in patients with IBD and increase in severity with active disease.⁵ These disorders may lead to poor compliance with medical treatment resulting in an increased morbidity rate.^6,7 A recent meta-analysis demonstrated an overall pooled prevalence of anxiety symptoms of 32.1% [95% CI 28.3-36.0] and of depressive symptoms of 25.2% [95% CI 22.0-28.5]. Patients with CD had a higher risk of anxiety and depression than patients with UC [OR 1.2, 95% CI 1.1-1.4].⁸ The Hospital Anxiety and Depression Scale [HADS] was introduced about 40 years ago as a self-assessment scale to aid the identification of mood disorders in the hospital setting.⁹ It was developed for the non-psychiatric patients and is the third most used self-screening tool for psychological outcomes.^10,11

A meta-analysis in 2009 concluded that for major depressive disorders, HADS sensitivity and specificity were 0.82 [95% CI 0.73–0.89] and 0.74 [95% CI 0.60–0.84] for a cut-off point of 8 and 0.56 [95% CI 0.40–0.71] and 0.92 [95% CI 0.79– 0.97] for a cut-off point of 11, respectively.¹² In the meta-analysis by Barberio et al., a HADS-Anxiety [HADS-A] score ≥8 demonstrated a pooled prevalence of 33.2% [95% CI 29.3–37.2; p <0.0001] and a HADS-Depression [HADS-D] score ≥8 of 21.6% [95% CI 18.4–24.9; p <0.0001].⁸ Prevalence of symptoms using a HADS score ≥11 was 21.3% [95% CI 17.2–25.6] for anxiety and 10.6% [95% CI 8.6–12.8] for depression symptoms.⁸

Moreover, IBD patients experience a significant socioeconomic burden and frequent periods of work disability, with a total yearly direct health care cost estimated at 4.6-5.6 billion euros in Europe.⁴ The economic impact is even higher, since IBD is affecting patients at an early age.^4,13 Absenteeism and unemployment are the most obvious states of low productivity. Presenteeism, which represents ongoing physical and/or mental health conditions preventing employees from being fully productive at work, has emerged as an indirect parameter affecting costs even in a higher rate than absenteeism.^14–16

Presenteeism rates vary in the general population from 30% to 90% due to different methods of assessment.¹⁷ Costs of presenteeism in IBD are difficult to quantify and there are limited published data. A recent study of 320 IBD patients from Finland estimated the mean annual economic cost of presenteeism to be 643,90 €/patient, with even higher costs in women [801.5 €/patient].¹⁸ More than one-third of IBD patients reported low presenteeism in a study by Shafer et al., and this showed a strong correlation with disability, low quality of life and emotional distress.⁹

Meanwhile various studies have explored the role of exercise in chronic medical conditions. Most of these demonstrated that engagement in physical activity may lower anxiety and depression states.^16,19–22 However, published data are scarce regarding the inter-relationship of all these aforementioned parameters in the everyday life of IBD patients.

The primary aim of this study was to determine the prevalence of anxiety and depression in IBD patients while in remission, using the HADS score. The secondary aims were to identify: [1] the prevalence of presenteeism at work; [2] the clinical characteristics associated with a greater incidence of psychological disorders; [3] the frequency of exercise engagement; and [4] any potential association of all these parameters with regards to daily status of the IBD patients. Ethical approval for this study was obtained from the university research and ethics committees.

2. Method

This was a prospective study, where IBD patients from eight European centres and Israel were recruited. The inclusion criteria were: age ≥16 years, and in clinical remission with no use of corticosteroids, flare-ups or complications in the p12 months. Remission was evaluated by the gastroenterologist collecting the data and was based on clinical scores [Crohn`s Disease Activity Index for CD²⁴ and Mayo score for UC^25] at the same clinical visit as that at which the questionnaire was completed, as well as colonoscopic and radiological studies performed during the past 2-3 years when available.

Data collection was carried out from September 2020 through March 2021. Other clinical data included: disease duration, age at diagnosis, smoking status, gender, education level, employment status, work location, relationship status, extra-intestinal manifestations [EIMs], history of admissions to hospital in view of IBD complications, current and past medications and psychiatric history.

The validated HADS score was used as a screening tool for anxiety [HADS-A] and depression states [HADS-D] [Supplementary data]. The Stanford Presenteeism scale [SPS-6] was used as a measure of work productivity. It consists of six questions using a five-point Likert scale [from strongly disagree to strongly agree]. The scale has strong evidence for internal consistency, structural validity, and criterion-related validity. Scores can range from 6 to 30, with a higher SPS-6 score indicating a higher level of presenteeism, i.e., a greater ability to concentrate on and accomplish work despite health problems [see Supplementary data]. The Godin Exercise Score was used as a validated questionnaire on physical activity. It defines three levels of fitness done over a period of 1 week, with scores adding up to three categories of exercise: insufficiently active/sedentary, moderately active, and active [Supplementary data]. The questionnaires used for the three scoring systems were translated into the local language in Italy, Portugal, Israel, Athens,and Piraeus. A forward and backward translation method was used for the process. In Malta, the questionnaires were given in English since it is one of the official languages. In the other centres in Greece, they were given in the English language.

Statistical analysis was done using the IBM SPSS Version 27. The chi square test was used to analyse categorical variables. Logistic regression analysis was used to relate a single dependent categorical variable to a number of risk factors. A forward stepwise procedure was used to identify the Parimonius model, which includes solely the significant predictors. From this analysis the odds ratio [OR] and corresponding 95% confidence intervals [CI] were extracted.

3. Results

3.1. Study population

A total of 585 patients with IBD were recruited. The majority of patients had CD [62.2%] and were male [53.0%], with a median age of 39 years [IQR 30-49]. Their clinical characteristics are demonstrated in Table 1. Three out of four participants [74.5%] were employed at the time of filling in the questionnaire. Table 2 demonstrates their level of education, relationship status, and use of illicit drugs.

Table 1.

Clinical characteristics.

Characteristics	Total
Age, median [IQR] years	39 [30-49]
Smoking history
Never smoked	48.0%
Ex-smoker	30.1%
Current smoker	21.9%
Montreal Classification
UC E1: proctitis	14.80%
UC E2: left sided	27.60%
UC E3: extensive disease	57.70%
CD non-stricturing, non-penetrating	54.80%
CD non-stricturing, non-penetrating + perianal disease	18.70%
CD stricturing	11.80%
CD stricturing + perianal	7.50%
CD penetrating	2.50%
CD penetrating + perianal	4.70%
Perianal disease [total]	14.70%
Medication
Nil or 5-ASA only	16.90%
Immunosuppressants [thiopurines or methotrexate]	7.30%
Biologics	60.90%
Biologics + immunosuppressants	14.90%
Length of diagnosis, median [IQR], years	9 [12]
Previous history of hospitalisation for IBD	62.2%
EIMs	23.60%

IQR, interquartile range; 5-ASA, 5-aminosalicylates; IBD, inflammatory bowel disease; EIMs, extra-intestinal manifestations.

Table 2.

Level of education, relationships, drugs of abuse.

Employment status
Employed	74.50%
Unemployed	25.50%
Relationship status
Single	32.7%
With partner	67.3%
Level of education
Primary	6.00%
Secondary	18.40%
High school/6th form	24.40%
Trade/vocational	8.60%
Undergraduate	11.70%
Postgraduate	31.00%
Drugs of abuse
None	91.20%
Marijuana	8.40%
Cocaine	0.40%

3.2. Prevalence of anxiety and depression

A quarter [25.0%] of our cohort had a formal diagnosis of anxiety or depression. This was commoner in CD [27.5%, p <0.05], in CD patients with stricturing and perianal disease [p <0.023], in females [29.1%, p <0.05], and in patients with a longer duration of disease [p <0.05]. Prior to the diagnosis of IBD, a diagnosis of anxiety and/or depression had been made in 10.8% of patients. Following the diagnosis of IBD, a further 14.2% of patients were diagnosed with these mood disorders. Assessment by a psychiatrist or psychologist occurred in 37.6% of patients, this being more frequently seen in patients with CD [41.6% of CD, p <0.05]; 11.8% of patients were currently on psychiatric medication. Use of such medications was more frequent in CD patients [14.7% of CD, p <0.01] and in single patients when compared with those who had a partner [17.3%, p <0.01]. Patients using drugs of abuse were more likely to have a diagnosis of depression or anxiety than non-users [39.2%, p <0.05]. On regression analysis, female gender [OR 1.91, 95% CI 1.01-3.60; p <0.05] and longer IBD duration [OR 1.04, 95% CI 1.01-1.07; p <0.01] were identified as risk factors for a diagnosis of depression or anxiety made by a clinician.

3.3. HADS score

An elevated HADS score [≥8, HADS-A or HADS-D] was present in 46.1%. Associations with IBD diagnosis [UC or CD], Montreal Classification, hospitalisation for IBD, age at diagnosis, or duration of diagnosis were analysed and commented on when significant. There was no significant difference between CD and UC patients regarding both scores. Patients who used recreational drugs were more likely to have an elevated HADS-A or HADS-D score [≥8] than non-users [62.7% vs 44.7%, p <0.05].

3.4. HADS-anxiety

HADS-Anxiety [score ≥8] was elevated in 41.8% of patients and in 25% of patients the score was ≥11. Patients with UC proctitis [E1 disease] were more likely to have an elevated HADS-A ≥11 [p <0.05] compared with patients with more extensive UC disease, as were patients diagnosed between 16 and 25years compared with patients diagnosed later on in life above the age of 36years [29.7% vs 19.7%, p <0.05].

3.5. HADS-depression

HADS-Depression was elevated in 24.5% [score ≥8] and 8.1% of patients had a score ≥11.

3.6. HADS regression analysis

Females more often had an elevated HADS-A [OR 1.89, 95% CI 1.17–3.08, p <0.01] or a HADS-D [OR 2.18, 95% CI 1.07-4.43, p <0.01] score ≥8 compared with males. The overall OR for a female patient to have either a HADS-A or a HADS-D score ≥8 was 1.7 [95% CI 1.1-2.8]. Patients with perianal disease were more likely to have a HADS-D score ≥8 [OR 2.20, 95% CI 1.07-4.43, p <0.05]. Finally, the risk factors for a HADS-A score ≥11 were female gender [OR 1.84, 95% CI 1.06-3.19, p <0.05] and treatment with a biologic agent [OR 2.62, 95% CI 1.10-6.28, p <0.05]. There were no factors associated with a HADS-D score ≥11.

3.7. Employment and presenteeism

In total, 74.5% of IBD patients were currently working and there was a difference in employment rate between patients with UC and CD [79.9% and 71.3%, p <0.05]. Unemployed patients were more likely to have a HADS-A or a HADS-D score ≥8 [30.8% vs 20.8%, p <0.01]. Low presenteeism at work as defined by a SPS-6 score ≤18, was evident in 34.0% of patients. There was no difference in presenteeism scores between UC and CD, gender, medications administered, disease classification, presence of perianal disease, duration since diagnosis, or age at diagnosis. Patients with a working pattern demanding physical presence demonstrated lower presenteeism than those working remotely [36.8% vs 23.6%, p <0.05].

Patients with anxiety or depression showed lower presenteeism at work [37.5% vs 23.2%] either overall or with a HADS-A or a HADS-D score ≥8 [67.2% vs 41.0%, p <0.01] and a HADS-A or HADS-D ≥11 [49.2% vs 20.5%, p <0.001] in particular. Similar difference was also detected independently of the time of IBD diagnosis [13.3% vs 7.2%, p = 0.05 in pre-IBD and 24.2% vs 16.1%, p = 0.05 in post-IBD diagnosis patients]. It is also interesting that patients who tested positive for Covid-19 showed higher presenteeism at work.

3.8. Godin Exercise Score

Physical activity being within the active range using the Godin Exercise Score was present in 37.3% of patients. A sedentary/insufficient exercise score was present in 22.3% of patients, with the rest [40.4%] being moderately active. There was no difference in exercise scores between CD and UC; however, UC patients with proctitis were more active than patients with left-sided or extensive disease [p <0.05]. There was no significant difference between exercise patterns for medications administered, disease classification, presence of perianal disease, duration since diagnosis, or age at diagnosis. Patients with an increased HADS-A or HADS-D score were more likely to be sedentary, independently of the cut-off value used [p <0.05]. Low presenteeism at work was also associated with sedentary and insufficient exercise patterns [50% were sedentary, 29.4% active, and 20.6% moderately active, p <0.01]. On analysis of the patients receiving psychiatric medications, 49% reported a sedentary lifestyle, 27.7% an active exercise lifestyle, and the rest [23.1%] being moderately active.

4. Discussion

Depression and/or anxiety almost doubled from 10.8% before IBD diagnosis to 20.3% after a diagnosis of IBD. These values were the result of a formal diagnosis done by a medical practitioner. However, when the HADS questionnaire was used as a screening tool, our data demonstrated that 46.1% of patients had a HADS-A or a HADS-D score ≥8. This indicates that practically half of our cohort had evidence of clinical anxiety or depression. When using an even higher cut-off value of ≥11, which is highly indicative of a mood disorder, more than one-fourth [27.4%] of patients reported a high HADS-A or HADS-D, which is again higher than the overall diagnosis done by a physician. This indicates that a significant proportion of patients with IBD have mood disorders which are currently undiagnosed and thus undertreated. No significant difference was seen in the results between the different centres.

Our cohort demonstrated higher rates of anxiety [41.8% vs 33.2%] and depression [24.5% vs 21.6%], using a cut-off value of ≥8 in the HADS score, when compared with the recent meta-analyses by Barberio et al.⁸ This may be explained by the fact that our study was performed during the Covid-19 pandemic which, most undoubtedly, could have led to higher levels of mood disorders in patients. On the contrary, HADS-A or HADS-D scores ≥11 were similar [27.4%] to those of the meta-analysis [31.9%], indicating that more serious psychiatric disorders are not influenced to the same degree by external non-mental factors. However, it is important to note that the meta-analysis included a mixed population of patients with active and quiescent IBD.

Risk factors for an elevated HADS-A score were female gender, use of biologics, and longer IBD duration. Surprisingly, patients with proctitis had a higher HADS-A score than patients with more extensive disease. This might be due to patients being more symptomatic with, for example, mucus per rectum or urgency. An elevated HADS-D score was associated with female gender and perianal disease. The presence of these findings confirms the validity of our patient cohort. Furthermore, patients who made use of illicit drugs were more likely to have an elevated HADS score.

Limited data are available on presenteeism at work in IBD patients. Interestingly, a higher rate of presenteeism was detected in IBD patients compared with controls [62.9% vs 27.3%, p <0.004] despite 35.6% of IBD patients being unemployed.²² In our study 74.5% of patients were working, with more patients with CD than UC being unemployed. A high SPS-6 score was present in 66% of patients, reflecting a satisfactory rate of presenteeism. This rate was lower in patients working physically on site than in those working remotely. This difference could be a consequence of added stress from being obliged to keep a physical presence at work and maintaining an increased risk of acquiring Covid-19. The SPS-6 score asks patients to relay information on the past month of work. Interestingly, patients who tested positive for Covid-19 showed higher presenteeism at work. A possible explanation could be their thoughts of potential repercussions at work and stigma of having Covid-19. A lower presenteeism at work was significantly associated with patients having an elevated HADS score, reinforcing the strong relationship between mood disorders and work productivity.

Regular exercise can improve psychological health by reducing depression and anxiety. This is relevant to IBD patients who have additional stress and impaired quality of life compared with the general population.^20,21 In a study by Chan et al. where 66% of IBD patients were currently participating in some form of exercise, 72% of the responders answered that exercising made them feel better and improved their general well-being and confidence.¹⁹ In another clinical trial by Justesen et al., employees who were randomised to exercise, within working hours and also off work, significantly improved presenteeism with a 6% increase in productivity.²³ In our study, 22.3% of patients had a sedentary lifestyle. Patients with a HADS-A and a HADS-D score ≥8 were more likely to have a sedentary lifestyle. The effect of the psychological disorder may thus limit the desire for physical activity. Furthermore, our study demonstrated that patients with sedentary or insufficient exercise had significantly lower presenteeism at work [p <0.01], again reinforcing the strong relationship between physical status and work productivity.

Our study has some limitations. The questionnaires with the scoring systems were translated into Italian, Portuguese, Greek, and Hebrew using a backward and forward translation method. This still raises some concern about the validity of score data, given they were not set in the original language. The scores were performed at a specific time point and these could vary with time. In terms of presenteeism, the SPS-6 score enquires about work productivity over the past month of work. This may vary with time; however, it should decrease the study recall bias. However, to limit confounding factors, we only recruited patients who were in clinical remission, with no use of corticosteroids, flare-ups, or complications in the previous 12 months. Furthermore, this was a multicentre study where a significant number of IBD patients in remission were recruited. This should give an overall picture of the prevalence of the mood disorders, despite individual Covid-19 issues in different countries.

In conclusion, our study demonstrates that in our cohort of 585 IBD patients in remission, a significant percentage suffer from anxiety and/or depression. One out of four patients was formally diagnosed with depression or anxiety, but 46.1% of them scored positive for a mood disorder on the HADS questionnaire. In this study we also identified risk factors for these mood disorders, and therefore greater attention should be given to at-risk groups such as CD patients, females, patients treated with biologic medications, and those exhibiting perianal disease. Furthermore, these elevated psychological scores were associated with unemployment, low presenteeism at work, poor physical exercise, and use of illicit drugs.

Rather than relying on the patient to come forward and seek help, the physician should familiarise herself or himself with these validated screening tools in order to increase the diagnostic rate of such pathologies and enable better holistic care for the IBD patients. Active involvement of a psychologist and/or a psychiatrist, as part of the IBD team, should be pursued to further improve the patients’ quality of life, which has emerged as one of the top priority outcomes in IBD.

The authors confirm that the data from this study are available within the article and the Supplementary material. Further raw data in support of the findings are available from the corresponding author.

Funding

None.

Conflict of Interest

FF: Janssen. KK: speaker fees from Abbvie, Aenorasis, Genesis, Galenica, Janssen, MSD, Pfizer, and Takeda and consultancy or advisory board member fees from Abbvie, Amgen, Ferring, Genesis, Janssen, MSD, Pfizer and Takeda; other speakers, see updated list on ECCO website.

Author Contributions

MS, PE: literature search, study design, writing first draft of article. MS, LC: data analysis and statistics; LL, CN, TT, PN, AT, AM, IG, HY, HA, GJM, TG, KF, IK, MA, GF, FF, KHK, FF, GM, JT: patient recruitment, data collection, review of article, and final approval of article for publication. Conference presentation: European Crohn’s and Colitis Organisation meeting, ECCO 2021; United European Gastroenterology Week, UEGW 2021.