Explaining fatigue: An examination of patient causal attributions and their (in)congruence with family doctors’ initial causal attributions

Abstract

Background: General practitioners (GPs) and patients can have different ideas about the causes of fatigue, which may hinder management of fatigue.

Objective: To investigate the causal attributions of patients and their GPs for fatigue, their level of agreement, and the association between patients’ attributions, and fatigue characteristics and other illness perceptions.

Methods: Baseline data, collected between 2004 and 2006, of a prospective cohort study among 642 adult patients presenting to Dutch primary care practices (n = 147) with a main symptom of fatigue, were used. Patient causal attributions and illness perceptions were measured using the revised illness perception questionnaire (fatigue version). GP causal attributions were measured with an open question included in the form that was completed at the end of the patient's visit. Fatigue severity was measured using the checklist individual strength.

Results: Psychosocial causes were among the most often reported causal attributions by both patients and GPs. In 33% of 519 cases, the GP had no idea about the cause whereas the patient did. Overall, the agreement between the first reported causal attribution of patients and GPs was low. Qualitative differences in the labelling of causes were also found. Type of attribution (physical vs psychosocial/psychological) was associated with duration of fatigue (40 vs 25 months), and personal control (score 17.4 vs. 18.9).

Conclusion: Most patients and GPs had ideas about the causes of fatigue, but differences were found in the first reported causes and the labelling of causes. The findings may provide leads for optimizing communication about fatigue.

Keywords::

• Anxiety
• depression
• somatization
• surmenage
• sleep
• general practice
• family medicine
• general
• communication

KEY MESSAGE:

• Illness perceptions—including causal attributions—play an important role in the course of fatigue.

• Patients and GPs mostly mention psychosocial causes for fatigue.

• This study shows that patients and GPs frequently have different ideas of perceived causes, indicating the need for explicitly addressing this issue in the consultation.

INTRODUCTION

Fatigue is a non-specific problem commonly presented in general practice. Fatigue can be experienced mentally and physically, it can vary in duration and severity ranging from mild fleeting fatigue, which everybody experiences at some point in time, to chronic severe fatigue (1,2), and it can also be verbalized in different ways. Fatigue co-occurs with many other symptoms and with many diseases while often the cause remains medically unexplained (1,2).

General practitioners (GPs) and patients often have different ideas about the causes of fatigue. Many studies have shown the importance and interrelation of illness perceptions in health and healthcare seeking (3–8). The understanding of such perceptions is essential for effective patient management and is compatible with an active role for the patient (9).

Causal attributions of patients, i.e. causes that patients ascribe their symptoms to, have been found to influence the course and severity of fatigue and the number of reported additional symptoms (10–13). GP's causal attributions about fatigue complaints partly determine how the consultation evolves (14). Underlying concerns of patients are often not clearly expressed and GPs often deal with expressed concerns in a medical way only, while patients may experience various problems such as depression, sadness, anxiety, work problems, family problems or other forms of stress (15). Emotional or psychosocial problems may be underreported due to patient expectations of the consultation, lack of time, embarrassment or no perceived possibility for receiving help (16). From the GP's perspective, (psycho)social causes may be difficult to define and manage (17). Patients experiencing persistent medically unexplained symptoms may be confronted with conflicting illness beliefs leading to tension in the doctor–patient interaction (18). Such disagreement in causes of fatigue may hinder communication and management of fatigue. Therefore, it is important to gain insight into the causes to which patients and GPs attribute fatigue complaints.

In this study, we aim to investigate causal attributions for fatigue of both patients and their GPs. The primary questions are:

(a)	To what causes do patients presenting with a new episode of fatigue attribute their fatigue?
(b)	What ideas do GPs initially have, and how well do these ideas match with patients’ beliefs?
(c)	Is the type of attribution reported by patients related to characteristics of fatigue and other illness perceptions?

METHODS

Recruitment of participants

Between 2004 and 2006, a cohort study was conducted among adult patients who consulted their GP because of fatigue in 147 practices across the Netherlands. Fatigue was defined as complaints of tiredness or synonyms indicating fatigue (e.g. exhaustion). Inclusion criteria were (a) presenting with a new episode of fatigue as a main symptom, i.e. not having visited the GP for fatigue in the previous six months, nor previously for the same episode of fatigue; (b) not being pregnant or treated for a malignancy; and (c) sufficient mastery of the Dutch language to complete the questionnaires. Of 856 eligible patients, 642 gave informed consent and participated in the study. They completed a questionnaire shortly after the baseline consultation and after one, four, eight and 12 months, respectively.

We used baseline data from this cohort study to answer the current research questions. At baseline, most participants (58%) had fatigue complaints for at least six months, mostly with a gradual onset (82%) and the average fatigue severity score at baseline was 46 on the checklist individual strength (see below for additional information). Many patients had elevated levels of distress, other symptoms/somatization and sleep problems (19).

Ethics

The medical ethics committee of VU University Medical Center approved the study (number 2004/04228, June 2004).

Measurements

Fatigue. Fatigue was assessed with the subjective fatigue subscale of the checklist individual strength (CIS) (20). This subscale consists of eight items that are scored on a seven-point scale. Scores can range from eight to 56, with higher scores indicating higher levels of fatigue severity.

Fatigue duration (number of years, months or weeks) and onset (sudden or gradual) were measured with self-report questions. Localization of fatigue was measured using multiple-choice options, which were grouped for analysis into: fatigue experienced ‘only in extremities,’ ‘only in the head’ and ‘in head and extremities or the whole body.’

Patient attributions and illness perceptions. Illness perceptions regarding fatigue were measured using the fatigue version of the revised illness perception questionnaire (IPQ-R) (21). Causal attributions are measured in two ways: using a standardized list containing 18 potential causes, and with an open question to rank-order (three) causes that are most important. These self-reported causes may come from the standardized list or be additional causes. The self-reported causes were used for analysis in this study. Other illness perceptions were measured using the following subscales from the IPQ-R adapted for fatigue: ‘timeline’ (patient expectations of the future course; six items), ‘personal control’ (self-efficacy; six items), and ‘illness coherence’ (five items). Items are scored on a five-point scale and summed to produce continuous subscale scores.

GP attributions. At the initial enrolment of patients in the study, GPs were asked whether they had any idea about the cause(s) of their fatigue (no; yes, possibly; yes, quite certainly) and if so, to specify their ideas.

Analysis

Two of the authors (IN, SL) independently categorized the patient self-reported causal attributions using the program Eyes & Archive, taking into account both specificity (homogeneity of reasons within a group) and frequency (do the amount of reasons warrant a unique category) of the causes. Categories were compared, and any differences discussed until consensus was achieved. If necessary a third investigator was consulted (HvdH). We applied the same method to the GPs’ ideas about the cause(s) for fatigue.

We reduced the initial specific categories to the following commonly used categories that could be used for comparisons and descriptive analysis: physical, psychiatric/psychological, psychosocial, psychosomatic/functional, behaviour/lifestyle, external. Supplementary file 1 (available online at http://dx.doi.org/10.3109/13814788.2015.1055556) provides a comprehensive overview of patients’ and GPs’ reported causes.

To answer our third research question, three main groups of categorized reasons given by patients (physical attribution only, psychological-based attribution only, or mixed attributions) and associations with fatigue characteristics and illness perceptions were analysed using Chi-square tests for dichotomous variables (localization of fatigue, onset) or t-tests for continuous variables (severity, duration, illness perceptions).

RESULTS

Patient attributions

Of the 642 patients, 567 (88%) reported at least one cause for their fatigue (Table 1). Of these patients, 472 (83%) patients reported two or more causes, and 343 (61%) reported three causes.

Table 1. An overview of the type causal attributions for fatigue provided by GPs and patients in a cohort of patients presenting to the GP with fatigue (n = 642 cases).

	Type of causal attributions provided by the GP	Type of causal attributions provided by the patients
Only physical	92 (14%)	80 (14%)
Only psychological	37 (6%)	6 (1%)
Only psychosocial	162 (25%)	88 (14%)
Only behaviour/lifestyle	2 (0.3%)	6 (1%)
Only external	0 (0%)	2 (0%)
Mixed attributions	83 (13%)	362 (56%)
Including a stated physical reason	41	173
Psychosomatic or functional	17	4
Don't know/fatigue NOS	208 (32%)	23 (4%)
Missing	58 (9%)	75 (12%)
Total	642 (100%)	642 (100%)

Psychosocial and physical causes were most often firstly mentioned while behaviour, lifestyle or psychological causes were more often mentioned secondly or thirdly (Table 2). A minority of patients reported only a physical cause or stated they had no idea about the cause.

Table 2. Number of causes mentioned by patients who reported at least one causal attribution for their fatigue (n = 567) categorised in order of importance (i.e. top three causes).

	n causes (%) as categorized in order of importance
	First cause	Second cause	Third cause
Physical	205 (36)	127 (27)	92 (27)
Psychological	42 (7.4)	55 (12)	59 (18)
Psychosocial	258 (46)	217 (46)	118 (34)
Psychosomatic or functional	2 (< 1)	2 (< 0)
Behaviour/lifestyle	25 (4.4)	59 (12)	50 (15)
External	9 (1.6)	7 (1.5)	17 (4.6)
Don't know/fatigue NOS	26 (4.6)	5 (1.1)	7 (2.0)
Total	567	472	343

GP reports of causes

For 636 patients, GPs reported whether or not they had an idea about causes for the fatigue. GPs reported to have no idea in 194 cases (30%) and to have an idea in 442 cases (70%); 259 times ‘possibly’ (41%) and 183 times ‘quite certainly’ (29%).

GPs specified their idea in 390 cases. GPs mostly reported one cause (290 cases, 74%). For 88 patients (22%) they reported two causes and for 12 patients (3%) three causes. Psychosocial causes were most often reported. Physical causes with no report of other causes were reported in 14% of cases and a combination of causes in 13% of cases (Table 1).

Qualitative analysis of patients’ and GPs’ causal attributions

A qualitative analysis of the causal attributions showed differences in how explanations of fatigue were articulated by patients and GPs (Supplementary file 1 available online at http://dx.doi.org/10.3109/13814788.2015.1055556). Musculoskeletal problems or pain was often mentioned by patients, but less by GPs who mainly reported (rheumatic) disease rather than complaints. Thyroid problems and anaemia were more often reported by GPs than patients while patients mentioned metabolic problems (diabetes, menopause) and pulmonary problems more often. Patients and GPs articulated psychological and psychosocial causes differently. GPs more often reported depression and illness labels such as overload or burnout while patients frequently mentioned reasons for overload such as stress or work-related problems.

Comparison of first mentioned causes by patients and GPs

The method of eliciting causal attributions was different for patients and GPs—patients completed a given list of causes before and were explicitly asked to report multiple causes (if present) while GPs were asked to report ‘what could be the matter’ with the patient. Results showed that patients often reported multiple causes (83%) while only 140 (36%) of the 390 GP reports mentioned multiple causes. Therefore, we decided to use only the first reported ideas of both GPs and patients for comparison (Table 3).

Table 3. A comparison of the first reported causal attribution of patients (rows) and GPs (columns), for the same episode of fatigue (i.e. patient–GP pairs; n = 516)^a.

Patient first reported causal attribution for the episode of fatigue (rows)	GPs first reported causal attribution for the episode of fatigue (columns)
	Physical	Psychological	Psychosocial	Behaviour/lifestyle	External	Syndrome diagnosis or ‘body and mind’	Don't know/fatigue NOS	Total (patients)
Physical	53^b	10	41	0	0	5	73	182
Psychological	6	8^b	13	0	0	1	13	41
Psychosocial	33	20	102^b	1	0	10	70	236
Behaviour/lifestyle	4	2	7	1^b	0	0	10	24
External	0	0	1	0	0^b	0	6	7
Syndrome diagnosis or ‘body and mind’	0	0	0	0	0	0^b	1	1
Don't know/fatigue NOS	10	2	3	1	0	0	9^b	25
Total (GPs)	106	42	167	3	0	16	182	516

^a126 cases were missing, from either GP or patient questionnaire.

^bNumber of matching causes.

In 34% of cases, there was a similarity in causes first reported by patients and GPs; 10% concerned a physical cause and 20% a psychosocial cause. When the GP did not know the cause, patients mostly had a clear idea (34% of all cases). The extent of the agreement was low with a kappa of 0.120.

Patient causal attributions and fatigue characteristics, and other patient perceptions

The type of attribution (only physical vs psychosocial/psychological/lifestyle) reported by patients was related to duration of fatigue and perceived personal control. Patients with only physical attributions had a longer duration of fatigue (40 vs 25 months, P = 0.034) and less personal control (score 17.4 vs 18.9, P = 0.001) compared to patients with other attributions. Type of attribution was not related to other fatigue characteristics or other patient perceptions. Patients who did not know the cause of their fatigue more often reported a sudden onset of their fatigue (36% vs 17% for the total group; P = 0.065).

DISCUSSION

Main findings

Most patients and GPs reported at least one causal attribution for the presented fatigue complaints. Both patients and GPs frequently reported psychosocial causes, while a minority in both groups reported a physical cause only. A comparison of the first-reported causal attribution of patients and GPs, for the same episode of fatigue, revealed that the ideas about the cause of fatigue matched in 34% of cases. Interestingly, in another 34% of cases the GPs indicated to have no idea about the cause of fatigue whereas the patient did. Overall the agreement between GPs and patients for the first-reported causal attribution was low. A more in-depth qualitative look at how the causes of fatigue were articulated showed differences between GPs and patients in the type of illness reported and the label given to the cause. Finally, we found that patients who reported a physical causal attribution tended to have a longer duration fatigue, and fewer feelings of personal control as compared to patients who reported psychosocial/psychological/lifestyle attributions.

Strengths and limitations

This study provides a quantitative and qualitative analysis of causal attributions for fatigue in a large representative sample of patients and their GPs. Although the data are derived from a study carried out between 2004 and 2006, there are no recent studies, new guidelines or other publications suggesting that causal attributions of patients or GPs on fatigue may have changed in the meantime. Therefore, we assume that our findings are still valid and relevant.

The method of data collection differed for patients and GPs. Only a minority of GPs reported more than one cause. Although GPs were free to report any idea they had, explicitly asking the GPs to rank-order several causes, could have triggered them to list further possible causes. Moreover, patients completed the open-ended questions after checking off a list of potential causes for their fatigue. This might have biased their spontaneous answers. Therefore, we compared the first (i.e. main) causes reported by GPs and patients.

We used self-report questionnaires. Interviewing several patients and GPs could have supplemented and further clarified our findings regarding their attributions and communication.

Interpretation of study results in relation to existing literature

The fact that most patients and GPs reported at least one causal attribution for the presented fatigue complaints may be a positive finding since not being able to make sense of symptoms can be very distressing for people (22).

Contrary to this study in which 56% of patients reported a combination of physical and other causal attributions (including psychological causes), results of an earlier study showed that most fatigued patients had physical attributions while most GPs considered a psychological cause; a minority of patient–GP pairs had mixed attributions (23).

Differences in first-reported causes mainly represented cases in which GPs indicated they had no idea while patients reported a psychosocial or physical cause. GPs could try to elicit these attributions, to address patients’ perceptions and worries more adequately. In half the cases in which patients reported a psychosocial cause first, their GP did not report a similar first cause. This may indicate that such causes are often not fully addressed in the limited consultation time, which may influence the consultation. When GPs have psychosocial attributions, consultations are characterized by more counselling and a longer duration, and less by medical examinations and the prescription of medicine (14).

Psychological and lifestyle attributions of patients were more often mentioned as a second or third cause in this study. This may indicate that personal attributions may be less easily reported and discussed by patients than psychosocial causes that are related to external situations. Alternatively, lifestyle was hardly mentioned by GPs as a cause for fatigue. This suggests that it is not discussed in consultations with patients presenting with fatigue, which is in line with previous studies (24). Unfavourable lifestyle patterns are commonly associated with fatigue and both under and over activity increase the risk of persistent fatigue (25–27). Adopting a health-enhancing lifestyle may be hindered by fearful beliefs about activity in chronically fatigued people (26). These results call for more attention for lifestyle patterns in consultations when patients present with fatigue.

Together with previous findings our findings underscore the importance of patient perceptions, including the type of attribution, with respect to expectations about the duration, future course of fatigue and the experienced personal control over fatigue (28,29). A longer fatigue duration may diminish the sense of control. Having psychological causal attributions seems to be favourable in terms of control and duration of fatigue, indicating an underlying sense of personal confidence and mastery regarding one's health. In patients with physical attributions, the sense of control over fatigue may be more dependent on the nature and perception of the physical problem (severity, chronicity, cause, diagnosis). Knowing that an existing disease causes fatigue does not exclude other (underlying) causes or maintaining factors, and equally calls for attention to the fatigue and subsequent (self) management (30–33).

Implications for clinical practice

Most patients and GPs had an idea about probable causes of the fatigue. While most patients and GPs (also) mentioned a psychological or psychosocial cause, these may often not be fully addressed. The association between psychological fatigue attributions and perceived personal control, points toward the potential benefit of exploring the patient's beliefs about their fatigue with an eye to empower patients and help them gain more insight into their complaints. The lack of causal ideas among one third of GPs, the lack of congruence of first reported causes, and qualitative differences in the articulation of causes may indicate that there may be room for improvement regarding communication or consensus about causes of fatigue. An important element of patient-centred care is finding common ground. Both consensus and lack of consensus may serve as a basis for further exploration of causes as well as for management possibilities in subsequent consultations or referral of patients presenting with fatigue.

Conclusion

Most patients and GPs had ideas about the causes of fatigue. While psychosocial causes were among the most frequently mentioned by both groups, overall the agreement between GPs and patients for their first-reported causal attribution was low. There were also some (qualitative) differences regarding the labels given to causes. The findings of our study may provide leads for optimizing communication about possible causes of fatigue in general practice.

SUPPLEMENTARY MATERIAL AVAILABLE ONLINE

Supplementary file 1 is available online at http://dx.doi.org/10.3109/13814788.2015.1055556

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.