OUP user menu

Journal of Crohn's and Colitis: 10 (8)


Laurence J. Egan, Ireland

Associate Editors

Shomron Ben-Horin, IsraelSilvio Danese, ItalyPeter Lakatos, HungaryMiles Parkes, UKJesús Rivera-Nieves, USABritta Siegmund, GermanyGijs van den Brink, NLSéverine Vermeire, Belgium


Published on behalf of

Personality and fatigue perception in a sample of IBD outpatients in remission: A preliminary study

Ingrid Banovic, Daniel Gilibert, Ahmed Jebrane, Jacques Cosnes
DOI: http://dx.doi.org/10.1016/j.crohns.2011.11.006 571-577 First published online: 1 June 2012


Background and aims: Fatigue is considered as a feature of IBD. Nevertheless, medical variables would partly explain this complex phenomenon. Psychological variables would be especially connected to fatigue for patients in remission. Moreover, personality is known to be linked to the fatigue of patients with CFS. This preliminary study aimed to determine if personality dimensions are linked to the perception of fatigue in IBD.

Methods: 81 IBD outpatients in remission completed the MFI (fatigue); ISI, EES (sleep disturbances); TCI-R (personality); HADS (depression and anxiety). Medical data were collected (ferritin, C-reactive protein, number of flare-ups, number of hospitalizations, duration of the disease and surgical sequelae).

Results: With the exception of surgical sequelae, none of the medical variables was linked to fatigue perception. Anxiety and sleep disturbances were the most continuously connected to fatigue perception. Significant relationships were observed between personality categorization on Persistence, Self-Directness and the level of fatigue.

Conclusion: In order to improve vitality in IBD patients in remission, identification and treatment of psychological aspects should become a dimension of disease management. Fatigue should not be considered only as a direct feature of IBD.

  • IBD
  • Fatigue
  • Personality
  • Depression
  • Anxiety
  • Sleep disturbances

1 Introduction

Fatigue in Inflammatory Bowel Disease (IBD) is a clinical feature of the disease. It is often considered to be a consequence of disease activity,1 anemia2 or side effects of the treatment.3 But when the disease is clinically inactive, a large proportion of the patients still suffer from fatigue (22 to 40%),4,5 suggesting that fatigue is related to other factors (medical or psychological) which could be important to identify to improve IBD patients' well-being.6 Quality of life studies710 have shown that active disease would involve a psychological distress that refers to depression and anxiety. Furthermore, fatigue could be a manifestation of depression11 or a consequence of anxiety.12 This psychological distress is often accompanied by sleep disturbances such as insomnia or hypersomnia.13,14 Sleep disturbances would be also linked to inflammation15,16 and to fatigue5 which underlines the necessity to consider them in IBD patients. Finally, the influence of personality in IBD related fatigue is little known even if the perception of fatigue could be modified by the effect of personality in other medical settings. Indeed, there is no real consensus concerning the role of personality in IBD. To our knowledge, only few studies17,18 evaluate personality (which is a relatively stable dimension over time for some of its dimensions) with specific tools. Most often, psychological variables (which are capable of evolving), such as coping strategies, anxiety or depression, which are indeed important, are evaluated.19,20,21 However, when personality was examined in IBD patients in order to determine its influence on quality of life,8 its potential links with fatigue were not evaluated, whereas personality factors have been shown to be related to the way in which individuals cope with their environment.22 Moreover, personality characteristics such as perfectionism23 would predispose the individual to develop or to maintain a fatigue state. In this perspective, personality and psychological distress have been assessed in chronic fatigue syndrome2426 or in other chronic diseases27 with the Revised Temperament and Character Inventory (TCI-R).28,29

Cloninger's biosocial theory of personality28,29 conceives personality as four basic biologically influenced dimensions that, in combination, would provide an underlying framework for both normal and pathological personality. Several studies have shown that the assessment with TCI-R would be pertinent to define personality dimensions as risk factors in somatic settings.8,27 In chronic fatigue syndrome, patients have been described as perfectionist (Persistence), neurotic (Harm-Avoidance) and quite obsessive (Persistence). Moreover, the more they had high scores in Self-Directness and low scores in Cooperativeness, the more they tended to show severe symptoms of chronic fatigue syndrome.24,25 Personality seems to be involved in this context in which no medical cause explains their fatigue. However, one wonders what possible link personality could have in a context in which the cause of fatigue is partly medically identified.5

Despite all these investigations, the potential relationships between personality dimensions and the fatigue of IBD patients have not been established, and a study including all the variables related to fatigue is lacking. Moreover, the physician's role in helping the patient may differ depending on which variable is most involved and which is remediable (such as gastrointestinal symptoms, sleep disturbances, and psychological suffering).5,9 By contrast, if personality is linked to fatigue perception, it would not be easily modifiable given that personality is a construct which remains relatively stable throughout life.27,28

For psychologists, IBD patients are pertinent for research because, in spite of a medical treatment for their disease1,6 and the fact that the disease appears inactive4,5, complaints of fatigue persist. Therefore, the aim of this preliminary study was to determine the relationships between personality and the perception of IBD related fatigue when the disease is in remission.

2 Material and methods

One hundred thirty-five patients were recruited at a French hospital when they had a consultation with their physician or when they had an infusion at the day treatment center. Patients were selected on the basis they complained of being tired. They were enrolled after an explanation of the study and they signed an informed consent form. From the initial sample, 81 ambulatory IBD patients were eligible (cf. Table 1): 40 patients were excluded because they had an active disease, one patient did not answer all the questions, and thirteen had a depressive mood (HAD-D > 8). Fifty-nine (73%) patients had been previously diagnosed with Crohn's disease (CD) and twenty-two (27%) with ulcerative colitis (UC).

View this table:
Table 1

Sociodemographic and disease-related data summarized for all patients.

Number of subjects81
Number of women38 (47%)
Age in years35.06 (± 11.74)
Marital status
Living as couple40 (49%)
Single35 (43%)
Divorced4 (5%)
No information2 (3%)
Education level
Primary secondary school2 (3%)
High school10 (12%)
9–11 years(CTE)10 (12%)
High school to master's degree28 (35%)
≥ Master's degree27 (33%)
No information4 (5%)
Conventional treatment
Anti-TNF α47 (58%)
Immunosuppressant + Anti-TNF α12 (15%)
Other16 (20%)
No treatment6 (7%)
Psychotropic treatment15 (18%)

2.1 Clinical and sociodemographic factors

Clinical and sociodemographic data included disease activity, age, gender, C-reactive protein (CRP), ferritin, current medical treatment, duration of the disease, number of prior hospitalizations (between 2006 and 2009 included), number of relapses (between 2006 and 2009 included), cumulative severity of intestinal resection assessed through the Post-Surgical Handicap Index (PSHI).30 This index is useful to predict the functional consequences following bowel resection, such as diarrhea and malnutrition. Disease history information was obtained from the MICISTA Registry.31 The Harvey–Bradshaw score32 for CD patients (score ≤ 4: inactive disease, score > 4: active disease) and the Lichtiger Index33,34 for UC patients were used to measure the clinical activity of the disease (score ≤ 6: inactive disease; score > 6: active disease). Relapses were defined as occurrence of any flare or complication leading to a change of IBD treatment.

2.2 Psychological variables and sleep assessments

The Multidimensional Fatigue Inventory was developed by Smets et al. (1995).35 The French version of the MFI36 has four dimensions (unlike the English version which has five): General Fatigue [GF] (i.e., the general sensation of being tired), Mental Fatigue [MF] (i.e., the cognitive aspects of fatigue), Motivation [Mo] (i.e. the difficulty to imagine enjoying a pleasant activity) and Reduced Activities [RA] (i.e., the patient's capacity to do physical activities). The score ranges between 9 and 45 for General Fatigue, 6 and 30 for Mental Fatigue, 3 and 15 for Reduced Activities, 2 and 10 for Motivation, and 20 and 100 for the MFI global score. The MFI allows one to make an overall estimate of fatigue (with a global score) and also to determine a profile of fatigue. Psychometric properties showed that the MFI is able to distinguish patients with different fatigue levels and it is sensitive to change (cf. Gentile et al., 2003).36

Depression and anxiety were assessed by the Hospital Anxiety and Depression Scale (HADS)37 in its validated French translation.38 This scale is designed for patients with physical illness so it does not include somatic symptoms of depression and anxiety. A score above 8 on either part of the questionnaire (HAD-A: anxiety, HAD-D: depression) indicates a probable clinical depression or anxiety.

The Insomnia Severity Index (ISI) provides an index of the global severity of insomnia including perceived daytime consequences and distress.39,40 Scores from 0 to 7 represent “no clinically significant insomnia”, scores from 8 to 14 represent “sub-threshold insomnia”; scores from 15 to 21 represent “clinical insomnia (moderate severity)” and scores from 22 to 28 represent “clinical insomnia (severe)”. The French version of the ISI is a reliable self-report measure to assess perceived sleep difficulties. It is able to detect changes in perceived sleep difficulties and its scores are convergent with those of the clinician.

The Epworth Sleepiness Scale (ESS) is a scale to assess daytime sleepiness.41 The subjects rate their probability of falling asleep on a scale of increasing probability from 0 to 3 in eight different situations. The score is the sum of the scores for the eight questions. A score in a range 0–9 is considered to be normal while a score in the range 10–24 is considered to indicate that a specialist physician's advice should be recommended. A score range of 9–14 is often considered as a sleep deficit and a score superior to 14 implies major daytime sleepiness. The EES has a high sensitivity and a high specificity.41

Personality was assessed with the Temperament and Character Inventory (TCI-R) in its validated French translation.29,42 It is a self-rating questionnaire which assesses the four dimensions of temperament and the three dimensions of character of the Cloninger psychological model of personality.43 The French psychometric properties are described by Pélissolo et al. (2000, 2005).29,42

This questionnaire assesses the 4 dimensions of temperament and the 3 dimensions of character of personality:

  • Harm-Avoidance [HA]: reflects a tendency to be shy, careful, passive, insecure and worried in anticipation of possible danger (the score range is between 33 and 165, median 99).

  • Novelty Seeking [NS]: expresses a tendency toward exploratory activity in response to novelty, impulsive decision-making, and active avoidance of monotony (the score range is between 35 and 175, median 105).

  • Reward-Dependence [RD]: expresses a tendency to have strong attachments, and to be sentimental and dependent on the approval of others (the score range is between 30 and 150, median 90).

  • Persistence [P]: reflects maintenance of behavior as resistance to frustration and fatigue; it correlates with rigidity and obsessiveness (the score range is between 35 and 175, median 105).

  • Self-Directness [SD]: refers to the ability to control, regulate and adapt one's behavior to fit the situation in accordance with chosen goals (the score range is between 40 and 200, median 120).

  • Cooperativeness [C]: refers to the tendency to be socially tolerant, emphatic, helpful and interested in other people (the score range is between 36 and 180, median 108);

  • Self-Transcendence: a trait associated with spirituality as well the ability to accept ambiguity and uncertainty (the score range is between 26 and 130, median 78).

2.3 Statistical analyses

The MFI global score provides an overall estimate of the intensity of fatigue. It is then possible, for a given sample, to determine which patients are the most and the less tired. Here, the score threshold is defined from the median (MFI median = 43) and allows the descriptive analysis.

Descriptive analyses were carried out in order to verify if:

  • there were differences on the psychological variables mean scores when patients were compared in function of their level of fatigue (Low Fatigue Level (MFI ≤ 43) and High Fatigue Level (MFI > 43)) using a nonparametric statistical procedure (Mann–Whitney U);

  • there were significant relationships between fatigue and medical treatment, the median number of flare-ups, the median number of hospitalizations, the PSHI, the median duration of the disease, and psychological and personality variables (considered here from a categorical approach), and a Chi-square was calculated.

Then, a Pearson's correlation was conducted with all the variables to test the intensity of the relation between the variables and the fatigue dimensions. Finally, multiple regressions were done with the medical variables and the psychological variables (that had a significant correlation with MFI global score; cf. Appendix A).

3 Results

3.1 Level of fatigue and inflammation and ferritin

The mean values of CRP were 7.71 (± 9.50) mg/L in the high fatigue group and 6.93 (± 7.09) mg/L in the low fatigue group. For ferritin, the mean values were 69.25 (± 57.05) ng/mL in the high fatigue group and 59.94 (± 56.44) ng/mL in the low fatigue group.

The comparison of means in function of the level of fatigue did not show significant difference for inflammation and ferritin (p >.05).

When the medical treatment type was compared between the patients dichotomized in function of the level of fatigue, the Chi-square test was not significant (p >.05). Fatigue perception was not linked to the medical treatment of these patients (Table 2).

View this table:
Table 2

Relationship between level of fatigue and medical treatment.

Low FHigh FChi-squared value
(MFI ≤ 43)(MFI > 43)
Anti-TNF α2819
Immunosuppressant + Anti-TNF α48χ2 = 8.19, p >.05
No treatment42

3.2 Level of fatigue and disease history

No significant relationship (p >.05) was observed (cf. Table 3) between the level of fatigue and the duration of the disease since the diagnosis, the number of severe relapses, the number of mild flare-ups, the number of quiescent periods, the number of hospitalizations (for major intestinal relapse) in the last three years.

View this table:
Table 3

Comparison of patients in function of level of fatigue and past relative disease information (number and severity of flare-ups, number of hospitalizations, cumulative severity of intestinal resection and duration of the disease).

Low FHigh FChi-squared values
(MFI ≤ 43)(MFI > 43)
Number of severe flare-ups (between 2006 and 2009)02012
11218χ2 = 3.14, p >.05
≥ 2109
Number of mild flares-ups03123
1814χ2 = 2.91, p >.05
≥ 232
Number of quiescent periods088
149χ2 = 3.05, p >.05
≥ 23022
Number of hospitalizations (between 2006 and 2009)None1812
One2021χ2 = 1.52, p >.05
≥ Two46
The cumulative severity of intestinal resection (PSHI)PSHI ≤ 2184χ2 = 4.79, p =.03
PSHI > 2128
Duration of the disease≤ 8 years2322χ2 =.02, p >.05
> 8 years1917

There was a significant relationship with the level of fatigue for the cumulative severity of intestinal resection (p =.03; median PSHI score = 21, IQ range: 17–31.5).

3.3 Fatigue and depression, anxiety and insomnia

The results in Table 4 show that patients in the high fatigue group suffered significantly more from anxiety (p =.01) and sleep disturbances (ISI: p =.004; ESS =.001) than patients in the low fatigue group. Moreover, there were significant relationships (cf. Table 5) between the level of fatigue and anxiety (p =.005), and daytime sleepiness (p =.01).

View this table:
Table 4

Mean scores in function of level of fatigue and U values.

Low FHigh Fp-value
(MFI ≤ 43)(MFI > 43)
Psychological assessmentsInsomnia (ISI)7.21 (± 4.44)10.25 (± 4.72).004
Daytime sleepiness (ESS)5.78 (± 2.90)8.67 (± 4.35).001
Anxiety (HAD-A)5.78 (± 3.26)7.54 (± 3.20).01
TCI-RNovelty Seeking103.78 (± 10.17)103.50 (± 11.72)ns
Harm-Avoidance94.64 (± 17.06)101.62 (± 17.72)ns
Reward-Dependence100.90 (± 13.47)99.61 (± 14.51)ns
Persistence125.59 (± 13.82)120.69 (± 18.60)ns
Self-Directness149.12 (± 13.48)137.69 (± 15.51).001
Self-Transcendence65.88 (± 11.91)70.49 (± 15.50)ns
Cooperativeness136.02 (± 13.21)130.58 (± 15.34)ns
View this table:
Table 5

Relationship between level of fatigue and depression (HAD-D), anxiety (HAD-A), sleep disturbances (ISI, PSHI, ESS) and personality dimensions (TCI-R).

Fatigue levelChi-squared values with 1ddl
Low FHigh F
(MFI ≤ 43)(MFI > 43)
AnxietyAnxious patients (HAD-A > 8)515χ2 = 7.67, p =.005
Not anxious patients (HAD-A ≤ 8)3724
InsomniaPatients suffering from insomnia (ISI > 8)1825χ2 = 3.66, p >.05
Patients without insomnia (ISI ≤ 8)2414
Daytime sleepinessPatients with sleep deficit (ESS > 9)310χ2 = 5.66, p =.01
Patients without sleep deficit (ESS ≤ 9)3927
Personality dimensionsNovelty SeekingNS ≤ 1052722χ2 = 0.52, p >.05
NS > 1051517
Harm-AvoidanceHA ≤ 992318χ2 = 0.59, p >.05
HA > 991921
Reward-DependenceRD ≤ 90711χ2 = 1.56, p >.05
RD > 903528
PersistenceP ≤ 10538χ2 = 3.08, p >.05
P > 1053931
Self-DirectnessSD ≤ 12018χ2 = 6.73, p =.009
SD > 1204131
Self-TranscendenceST ≤ 783629χ2 = 1.64, p >.05
ST > 78610
CooperativenessC ≤ 10813χ2 = 1.21, p >.05
C > 1084136

3.4 Fatigue and personality dimensions

In Table 4, a significant difference was observed for Self-Directness (p =.001). The patients with a higher fatigue level were less able to regulate their behavior to fit the situation in accordance with chosen goals than the less tired patients. Moreover, when personality dimensions were considered as categories (according to the median score for each subscale of the TCI-R (cf. Material and methods)), there was only a significant relationship (cf. Table 5) between level of fatigue and Self-Directness (p =.009).

3.5 Multiple regression including all significant variables

Finally, multiple regression analysis showed that 45% of the variance of the MFI was explained by the value of PSHI (r =.27, β =.005), daytime sleepiness (β =.36, p =.0009), Persistence (β = −.21, p =.02), and Self-Directness (β = −.22, p =.04).

Persistence and Self-Directness were the dimensions of personality that were linked the most to fatigue.

4 Discussion

The aim of this preliminary study was to determine if personality dimensions were linked to the intensity of the perception of IBD related fatigue for patients in remission. The results showed that the only medical variable linked to the intensity of the global score of fatigue was the cumulative severity of intestinal resection (PSHI). The results of the multiple regression analysis for the psychological variables showed that links exist between the intensity of fatigue and the level of anxiety, insomnia, and the SD dimension of personality. However, only the risk of daytime sleepiness and the personality dimensions of Persistence and SD were the most pertinent for predicting the intensity of fatigue. In other words, personality variables explain the intensity of the fatigue of IBD patients in remission.

First of all, the results concerning sleep disturbances (insomnia and daytime sleepiness) are consistent with research showing that IBD patients suffer from fatigue and sleep disturbances even if the disease is inactive.4,5,15,16,44,45 They also illustrate that fatigue is strongly related to anxiety.12,46 However, if it is not possible here to know the cause of their sleep disturbances and their anxiety, both of them require an adapted treatment. Finally, these results underline the necessity to verify the different possible configurations between sleep disturbances, anxiety, medical factors related to disease, personal life and fatigue.47

As expected, personality dimensions were associated with the intensity of fatigue perception. More particularly, Self-Directness and Persistence were connected to IBD fatigue perception as it was observed elsewhere.2326 If this is confirmed by the results of a longitudinal study in progress, it can be hypothesized that personality dimensions may be involved in the adjustment of patients to their state of fatigue as it has been inferred for the quality of life.8 The results in these studies are consistent with those observed for subjective quality of life (Qol) or CFS assessments concerning Self-Directness. In the general population, Self-Directness was linked to the perception of health and happiness.4851 In our sample, one can hypothesize that the patients who have such a personality trait manage to benefit from the fact that their disease is in remission48 and, despite its aleatory aspect, maintain some well-being. Therefore, a patient who is less prone to having negative thoughts or emotions, would not perceive fatigue intensively when the disease is clinically inactive.26

However, concerning Persistence, our results are in contradiction with those observed for subjective quality of life (Qol) or chronic fatigue syndrome (CFS).8,23,24 Globally, this dimension of personality involving the tendency to persevere in behaviors had been previously associated with a reward of relief, despite frustration and fatigue. In other words, because of this dimension of their personality, these patients would be more likely to not take their sensation of fatigue into consideration or to minimize the sensation. But more specifically, according to the studies mentioned above, Persistence has a negative impact on the subjective quality of life or contributes to fatigue in the context of CFS. In our research, the more the subjects are Persistent, the less they are tired. This can be explained by several factors. First, the scores for Persistence in studies in which it plays a negative role, are higher. It is likely that the overly strong expression of this personality dimension (as its weakness) refers to an unadapted functioning against fatigue (curvilinear relation). Moreover, in the context of IBD, the disease is changeable: the fact of being relatively persistent could be seen as an adaptive factor. Effectively, the patient maintains an active life despite his symptoms or the consequences of his disease. He would then resist the adversity of his disease and its consequences. In other words, in spite of a difference in the results observed for IBD Qol,8 Persistence could have a beneficial or a negative influence according to its degree (which requires further investigation)52,53 and would be helpful in a context of chronic disease evolution.

Finally, in future research, it would be pertinent to deal with aspects of the personality which can be more easily accessed by a psychological treatment such as coping strategies.54,55 Effectively, it would be important to determine the extent to which they could be limited by the subject's perceived level of fatigue and the extent to which they could moderate the links between personality traits52,53 and the perceived intensity of fatigue.

When dealing with patients in remission who perceive themselves as tired, both gastroenterologists and psychologists have to consider not only fatigue, but also psychological suffering and sleep disturbances which may respond positively to treatment and psychotherapy.5659 This multidisciplinary approach (psychological and medical) would have the advantage of improving these remediable factors and, subject to subsequent validation, would be able to potentially improve the fatigue of these patients as a result. If more important medical and psychological factors are at stake in medical care, the dimensions of personality must not be underestimated.5,52,53,60

6 Conflict of interest

There is no conflict of interest.


The MFI global score correlated with insomnia (ISI: r =.39), ESS (r =.44), HAD-A (r =.27) and Harm-Avoidance (r =.31), and was negatively associated with Persistence (r = −.31), Self-Directness (r = −.40), and Cooperativeness, (r = −.28).


  • 1 To verify if anxiety (HAD-A) and insomnia (ISI) were linked, we conducted a Chi-square test which proved that there is no significant relationship between them (p >.05). Insomnia here was not related to the presence of an anxiety state.

    The same test was done for daytime sleepiness: there is no significant relationship between the presence of an anxiety state and the state of daytime sleepiness (p >.05).

  • 2 Post-Surgical Handicap Index.


View Abstract