The association of personality traits with poststroke fatigue in daily life: An exploratory experience sampling method and cross-sectional study

ABSTRACT

Fatigue is a frequently occurring and persistent symptom after stroke. Many biological, psychosocial, and behavioural factors are associated with poststroke fatigue, but research into associations with personality traits is relatively sparse. In this study, we explored whether personality traits were related to poststroke fatigue measured with conventional fatigue questionnaires as well as experience sampling methodology (ESM). Twenty-four individuals with stroke completed 10 daily questionnaires about momentary (here-and-now) fatigue for six consecutive days using the mHealth ESM application PsyMate^TM. Further, they completed questionnaires assessing personality (NEO-FFI and LOR-T) and fatigue (FSS). Results showed that higher extraversion (ß = -.44, SE = .12, p = .001; 95% CI = -.67-.19) and optimism (ß = -.18, SE = .06, p = .007; 95% CI = -.30-.05) were associated with lower momentary fatigue. No association was found between neuroticism and momentary fatigue, but higher neuroticism (r = 0.531, p = .008, 95% CI = .160-.759; r = .574, p = .003, 95% CI = .245-.767) was associated with higher scores on the retrospective FSS scales. We conclude that personality traits differentially influence poststroke fatigue, but this also depends on the way fatigue is measured (with retrospective or with momentary measures). When functional gains are not in line with expected progress during the rehabilitation treatment of fatigue, it may be appropriate to take into account how person characteristics are related to momentary fatigue.

KEYWORDS:

• Fatigue
• experience sampling method
• stroke
• personality traits
• fatigue severity scale

Introduction

Fatigue after stroke is highly prevalent, with a significant impact on daily life (Levine & Greenwald, 2009). Fatigue is reported by 30% to 68% of the individuals who suffer a stroke (De Groot et al., 2003). Moreover, poststroke fatigue (PSF) negatively affects health outcomes such as neurological recovery (Choi-Kwon & Kim, 2011) and is associated with physical disability and lower quality of life (Chen et al., 2015). PSF is more common in female patients and in individuals who already reported fatigue or mood symptoms before the stroke incident (Cumming et al., 2018). A distinction is commonly made between primary fatigue - as a direct result of brain damage – and secondary fatigue – related to mental and somatic comorbidity or medication (Aarnes et al., 2020; Choi-Kwon & Kim, 2011; De Doncker et al., 2018; Schönberger et al., 2014). A distinction is also made between mental and physical fatigue (Hinkle et al., 2017). Further, fatigue is to be distinguished from fatigability, with the latter referring to objective decrease in cognitive or physical performance as a result of prolonged effort (Kluger et al., 2013). Kutlubaev and colleagues (2015) suggested another type of fatigue, called non-exertion fatigue (Kutlubaev et al., 2015). That is, fatigue that is not associated with exercise, household activities, or social activities but seems to be associated with personal traits such as coping styles (Jaracz et al., 2007) and personality traits (Kendler & Myers, 2010; Leandro & Castillo, 2010).

Personality traits are defined as a particular pattern of thoughts, feelings, and behavioural modes in a variety of situations, which are relatively stable over time and often predict behaviour (Costa & McCrae, 1987). Personality also gives insight into a person’s response to challenging disease experiences or other stressful life events (Schreiber et al., 2015). Personality traits have been linked to fatigue in different clinical populations, with neuroticism being the most frequently reported trait associated with fatigue symptoms (e.g., Besharat et al., 2011; Sindermann et al., 2018; Sugawara et al., 2005). For instance, research in breast cancer patients showed that the degree of neuroticism was related to fatigue severity, independent of demographic and clinical variables (Besharat et al., 2011; Poeschla et al., 2013; Sugawara et al., 2005; Vassend et al., 2018). Further, Stone and Richards (2001) found extraversion (as a personality trait), but not neuroticism, to be significantly associated with higher fatigue scores among patients with cancers of the prostate or breast. In addition, Sindermann et al. (2018) found that low extraversion significantly predicts physical fatigue but not mental fatigue in patients with multiple sclerosis. To our knowledge, there is only one study that has investigated the relationship between neuroticism and fatigue after stroke. Lau and colleagues (2017) investigated this relationship in Chinese individuals with stroke and found neuroticism to be associated with higher fatigue severity, independent of depressive symptoms. Another personality trait that has been related to fatigue is optimism (Allison et al., 2003; Chambers et al., 2012). Optimism has been defined as generalized expectations that good things, rather than bad, will happen in the future (Scheier et al., 1994). For instance, Schou-bredal and Tøien (2017) argued that optimism might act as a protective factor against fatigue in breast cancer patients.

Taken together, researchers have shown the influence of personality traits on fatigue in different clinical populations (Besharat et al., 2011; Conversano et al., 2018; Sindermann et al., 2018), but little research has been done on the role of personality traits in relation to fatigue after stroke. Moreover, most studies investigating the relationship between personality and fatigue have used traditional and often retrospective questionnaires to assess fatigue. Importantly, the way and format in which questions are presented may influence the (reliability of) answers that are given (Van Den Bergh & Walentynowicz, 2016). Retrospective questions (e.g., “How tired did you feel during the past 7 days?”) rely on information from episodic autobiographical memory, which is prone to biases in self-report and memory. Individuals with stroke often have cognitive problems such as slow information processing, reduced concentration, memory problems, and executive functioning problems (Hinkle et al., 2017). These cognitive problems affect the way information is stored and retrieved and may influence the accuracy of retrospective fatigue reports. Further, biases in self-report are related to the so-called remembering self (Conway & Pleydell-Pearce, 2000). This sort of information is based on the story that is made around the experience with a situation and address personal semantic memory, which is affected by the way a person considers himself or herself, determined by personality, history and background (Robinson & Clore, 2002). This means that storing information about a complaint as fatigue, is influenced by how a person perceives himself and the world.

A method less influenced by these potential sources of bias is momentary assessment. Momentary data-collection, known as the experience sampling method (ESM), is a structured diary technique that makes it possible to investigate symptoms such as fatigue in real-time, over time, and across different contexts (Delespaul, 1995; Stone & Shiffman, 1994). ESM enables a more comprehensive and ecologically valid understanding of the frequency, intensity and duration of symptom experience (Sohl & Friedberg, 2008). Momentary data-collection does not rely on episodic memory and is free from memory bias (Heron & Smyth, 2010). The feasibility and usability of ESM in individuals with stroke have been demonstrated in several studies (e.g., Jean et al., 2013; Johnson et al., 2009; Lenaert et al., 2019).

This study aimed to explore the relationship between general, mental and physical fatigue in daily life and personality traits in stroke patients. We investigated the relationship of personality with both momentary and retrospective fatigue reports. Given that no research has been conducted using momentary measures of fatigue in relation to personality traits, we had no a priori hypothesis about (differences in) momentary versus retrospective measures of fatigue in their relation to personality. However, based on earlier research with conventional (retrospective) fatigue measurements (Lau et al., 2017; Poeschla et al., 2013; Schou-bredal & Tøien, 2017; Sindermann et al., 2018; Vassend et al., 2018), we hypothesized that individuals high in neuroticism would report higher levels of fatigue and that individuals high in optimism would report lower levels of fatigue, irrespective of the method used to measure fatigue. Based on the findings of Sindermann and colleagues (2018), we also hypothesized a negative association between fatigue and extraversion.

Method

Participants

Participants were recruited between September 2016 and October 2017 in Zuyderland hospital in Sittard, Adelante Zorggroep rehabilitation centre in Hoensbroek, and the University Medical Center in Maastricht (Netherlands). Inclusion criteria were: 1. Diagnosis of stroke confirmed by a neurologist, 2. Receiving outpatient rehabilitation care, 3. Age above 17 years and have the capacity to give consent. 4. Good comprehension of the Dutch language. Exclusion criteria were: 1. No possession of or unable to use a smartphone, 2. Participation assessed as potentially too burdening based on clinical judgement, 3. Diagnosis of chronic fatigue syndrome or fibromyalgia or currently undergoing cancer treatment (self-reported). The Medical research ethics committee of the Maastricht University Medical Center approved the study (approval code: METC 16-4-101). All participants gave their written informed consent.

Measurements

Psymate^TM

PsyMate™ is a smartphone-based mHealth application developed by Maastricht University and Maastricht UMC+ (www.psymate.eu) for moment-to-moment assessment of daily life experiences. The PsyMate application was programmed to prompt participants with beep signals 10 times a day at random times during six days between 7.30 AM and 10.30 PM, with the restriction that beeps were separated by at least 15 min and no more than 270 min. The average interval was set to 90 min. After each beep signal, a digital questionnaire was presented about current fatigue, mental state and activities. Participants had 15 min to respond after each beep before the questionnaire disappeared. Items concerning fatigue, assessed level of physical, mental and general fatigue (i.e., “I feel tired” or “I feel physically tired” or “I feel mentally tired”) on a seven-point Likert scale ranging from 1 (not at all) to 7 (extremely). Whenever participants responded two or higher to “I feel tired”, they were also asked to respond to the statements “I feel mentally tired” and “I feel physically tired”.

Questionnaires

Personality was assessed with the NEO Five-Factor Inventory (NEO-FFI) (Hoekstra, Ormel & De Fruyt, 1996). This self-report questionnaire consists of 60 statements covering the 5 personality traits extraversion, neuroticism, openness to new experiences, agreeableness, and conscientiousness. Participants reach each statement on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree”. The NEO-FFI has been used in stroke patients, with the scale showing satisfactory internal consistency (α = 0.57-0.86) (Dwan et al., 2017).

To assess individual differences in generalized optimism, the Life Orientation Test-Revised (LOT-R) was used. The questionnaire consists of 10 items, with six items keyed in a positively or negatively direction (e.g., “I’m always optimistic about my future” or “I rarely count on good things happening to me”) as well as four filler items. Items are scored on a 5-point Likert scale, ranging from 0 = strongly disagree to 4 = strongly agree. A total score from the 6 items (items 1,4, 10 and items 3,7,9 are reverse coded before scoring) is calculated (by summing the items), which represents an overall optimism score. Scheier et al., 1994, found that the LOT-R has good internal consistency (Cronbach's alpha .78) and the test-retest reliability was .68 (4 months), .60 (12 months), .56 (24 months) and .79 (28 months), suggesting that the scale is stable across time.

The Fatigue Severity Scale (FSS or FSS-9) assesses fatigue severity (Krupp et al., 1989). The FSS-9 consists of nine statements (e.g., “I am easily fatigued”) rated on a 7-point Likert scale. The total score is calculated as the mean score per item. Scores range from 1 to 7, with scores of 4 or higher pointing to potentially clinically relevant fatigue. Lerdal and Kottorp (2011) showed that the first two items of the FSS-9 (i.e., “My motivation is lower when I am fatigued” and “Exercise brings on my fatigue”) did not demonstrate acceptable goodness-of-fit in a stroke population and that a version of the FSS-9 that omits these two items, the FSS-7, demonstrates more robust psychometric properties for poststroke fatigue (Lerdal & Kottorp, 2011). Therefore, we also calculated the total mean score per item for the FSS-7 for all subjects.

Procedure

After stroke, individuals following an outpatient rehabilitation programme were screened by the treating therapist on the inclusion and exclusion criteria and were given an information letter if they were deemed eligible for and interested in participating in the study. After participants had given their informed consent, a briefing session was planned with the researcher. In this session, the PsyMate app was introduced, which was then installed on the participant’s smartphone. Participants were guided through the app and were given time to practice with the ESM questionnaire. Subsequently, each participant completed the ESM questionnaires for six consecutive days (not including the day of the briefing session). Additionally, they were instructed to continue their daily lives as usual and to avoid adjusting their daily routines to the study in any way (e.g., by switching off their phones if they wished to rise later than 7.30 AM or go to bed earlier than 10.30 PM). After the PsyMate period, a one-hour debriefing session was planned during which participants completed the NEO-FFI, FSS and the LOT-R. The researcher then provided graphical ESM-derived feedback to each participant.

Statistical analysis

All statistical analyses were performed using SPSS for Windows, version 25.0. We expected to find moderate to strong (r ≥ .50) correlations between retrospective and momentary measures of fatigue. In order to detect these effects at 80% power and α-level of .05, we needed to recruit at least 23 participants. Descriptive analyses were conducted to assess response rates. ESM data have a hierarchical structure because of within-subject clustering: data from ESM questionnaires (level 1) are nested within individuals (level 2). Questionnaires for personality and fatigue are also level 2 variables in this hierarchical structure. Therefore, multilevel linear regression analyses were used to relate personality (level 2 variable) with momentary fatigue data (level 1 variable). Prior to analysis, the ESM variables’ range was transformed to range from 0–6 to obtain meaningful intercepts. Multilevel analyses were run using an unstructured covariance structure for the random effects.

Separate multilevel regression analyses were used with individual personality traits (NEO-FFI) as independent variables and general, mental and physical momentary fatigue as the dependent variable (ESM questionnaires). For the analyses with NEO-FFI as predictor, norm scores according to sex were used. Therefore, we only included age as covariate in these models. For the analysis with optimism (LOT-R) as predictor, both age and sex were entered as covariates in the model. An alpha value of <.05 was used for statistical significance. The models were adjusted for multiple comparisons (with six predictors), and therefore we set the significance level at a p-value of 0.008.

The associations between personality traits (NEO-FFI and LOT-R) and the total scores of the FSS-9 and the FSS-7 was calculated using Pearson correlation analysis. Because the Shapiro-Wilks test of normality was significant for all these variables, indicating deviation from a normal distribution, bootstrapped 95% confidence intervals are reported.

Results

Sample characteristics

A total of 30 patients were approached to participate in this study. Six participants were excluded from the analyses. Two of them withdrew from the study within two days after the start; two others did not complete the self-reported personality questionnaire. Two other individuals completed less than 33% of all ESM questionnaires, which is a criterion used for exclusion conform to ESM guidelines (Delespaul, 1995). The remaining 24 participants (14 female) had a mean age of 55.5 years (SD = 7.6, range: 34 - 68). At the time of the study, five participants were using antidepressant or mood-altering medication. All patients were following outpatient rehabilitation. Every patient included in the study was past the acute stage after stroke. The majority of patients were between 3–6 months after stroke. Participants responded to 42 out of 60 beeps on average (SD = 8, range: 21-55). The total number of beeps responded to across all subjects was 1013. Descriptive statistics of the ESM measurements and the questionnaires are shown in Table 1. The partial correlation between momentary mental and momentary physical fatigue corrected for the general fatigue was 0.597, p <.001.

Table 1. Descriptive statistics of the fatigue and personality variables.

	Mean	SD	Range
General fatigue (ESM variable)	3.83	1.78	1–7
Mental fatigue (ESM variable)	4.08	1.59	1–7
Physical fatigue (ESM variable)	3.70	1.59	1–7
FSS-9	5.22	1.04	2.67 −7.00
FSS-7	5.31	1.22	2.14 −7.00
NEO-FFI-N	5.42	2.59	1–9
NEO-FFI-E	5.42	2.03	2–9
NEO-FFI-O	4.70	2.35	1–9
NEO-FFI-A	5.90	1.78	1–8
NEO-FFI-C	5.18	2.51	1–9
LOT-R	13.62	4.79	6–21

FSS = Fatigue Severity Scale; NEO-FFI = NEO Five-Factor Inventory, N = Neuroticism, E = Extraversion, O = Openness, A = Agreeableness, C = Conscientiousness; LOT-R = Life Orientation Test-Revised; Experience sampling method = ESM

Association between personality traits and momentary general fatigue (ESM variable)

Table 2 provides the output of the multilevel regression models with different personality traits (NEO-FFI variables) as predictor and momentary general fatigue (ESM variable) as the dependent variable, controlling for age and gender for the model with LOT-R optimism as predictor. Significant associations were found for extraversion and optimism. The model with extraversion as predictor reveals a significant negative association with the level of momentary general fatigue (ß = -.44, SE = .12, p = .001; 95% CI = -.67 -.19), but age was not associated (ß = -.02, SE = -.37, p = .590; 95% CI = -.08 .05). In the model with optimism as the predictor, age and gender as covariates, optimism was significantly and negatively associated with the level of fatigue (ß = -.18, SE = .06, p = .007; 95% CI = -.30 -.05) but age (ß = -.003, SE = .04, p = .940; 95% CI = -.08 .08) and neither gender (ß = .13, SE = .48, p = .789; 95% CI = -.86 1.12) were associated with the level of fatigue.

Table 2. Associations between general fatigue (ESM variable) and personality traits.

	Level of general fatigue
	ß	SE	p	CI
NEO-FFI-N	.18	.13	.181	-.09, .45
Age	-.02	.04	.555	-.11,.06
NEO-FFI-E	-.44	.12	.001**	-.67, -.19
Age	-.02	-.37	.590	-.08,.05
NEO-FFI-O	.007	.14	.960	-.29,.28
Age	-.06	.04	.181	-.15,.02
NEO-FFI-A	-.28	.16	.090	-.61,.04
Age	-.03	.04	.407	-.11,.05
NEO-FFI-C	-.27	.10	.012	-.48, -.07
Age	-.05	.03	.146	-.12,.01
LOT-R	-.18	.06	.007*	-.30, -.05
Age	-.003	.04	.940	-.08,.08
Gender	.13	.48	.789	-.86,1.12

Dependent variable ESM level of fatigue *p < .008; **p < .001. CI = 95% bootstrapped confidence intervals; NEO-FFI = NEO Five-Factor Inventory, N = Neuroticism, E = Extraversion, O = Openness, A = Agreeableness, C = Conscientiousness; LOT-R = Life Orientation Test-Revised

Association between personality traits and the levels of mental and physical fatigue (ESM variable)

Table 3a en Table 3b provides the associations between momentary mental and physical fatigue and personality traits (NEO-FFI variables). In the model with mental fatigue as the dependent variable and extraversion as the predictor and age as covariate a significant negative association was found (ß = −0.431, SE = 0.108, p< .001; 95% CI = −0.654-0.209). Conscientiousness (ß = −0.288, SE = 0.085, p =  0.003; 95% CI = −0.465 −0.111) and optimism (LOT-R) (ß = - 0.173, SE = 0.053, p =  0.003; 95% CI = −0.283-0.063) were also significantly and negatively associated with mental fatigue. No significant associations were found between physical fatigue and personality traits.

Table 3a. Associations of the ESM variable mental fatigue with personality traits (N  = 24).

	Level of mental fatigue
	ß	SE	p	CI
NEO-FFI-N	.169	.120	.171	-.078,.418
Age	-.050	.039	.213	-.132,.031
NEO-FFI-E	-.431	.108	.001**	-.654, -.209
Age	-.040	-.028	.167	-.098,.018
NEO-FFI-O	-.020	.126	.871	-.282,.241
Age	-.079	.038	.051	-.159,.000
NEO-FFI-A	-.217	.144	.146	-.517,.081
Age	-.061	.035	.095	-.133,.011
NEO-FFI-C	-.288	.085.027	.003*	-.465, -.111
Age	-.071		.018	-.128, -.013
LOT-R	-.173	.053	.003*	-.283, -.063
Age	-.019	.034	.574	-.090,.051
Gender	.032	.420	.939	-.833,.898

Dependent variable ESM level of mental fatigue *p < .008; **p < .001. CI = 95% bootstrapped confidence intervals; NEO-FFI = NEO Five-Factor Inventory, N = Neuroticism, E = Extraversion, O = Openness, A = Agreeableness, C = Conscientiousness; LOT-R = Life Orientation Test-Revised

Table 3b. Associations of the ESM variable physical fatigue with personality traits.

	Level of physical fatigue
	ß	SE	p	CI
NEO-FFI-N	.172	.119	.164	-.074,.419
Age	-.005	.039	.887	-.086,.075
NEO-FFI-E	-.277	.126	.038	-.538, -.016
Age	-.010	-.333	.748	-.079,.057
NEO-FFI-O	-.025	.126	.840	-.286,.234
Age	-.034	.038	.383	-.113,.045
NEO-FFI-A	-.293	.138	.046	-.580, -.005
Age	-.009	.033	.782	-.078,.060
NEO-FFI-C	-.228	.092	.021	-.419, -.037
Age	-.028	.030	.349	-.091,.033
LOT-R	-.106	.057	.073	-.022,.010
Age	.008	.036	.829	-.067,.083
Gender	-.532	.449	.247	−1.456,.391

Dependent variable ESM level of fatigue *p < .008; **p < .001. CI = 95% bootstrapped confidence intervals; NEO-FFI = NEO Five-Factor Inventory, N = Neuroticism, E = Extraversion, O = Openness, A = Agreeableness, C = Conscientiousness; LOT-R = Life Orientation Test-Revised

Association between personality traits and retrospective fatigue ratings

Table 4 provides correlations between the total scores of the FSS-9 and FSS-7 on the one hand and the personality dimensions (NEO-FFI variables) scores on the other hand. A significant positive correlation was observed between neuroticism and retrospective fatigue. Neuroticism explains between 28% (FSS-9) and 33% (FSS-7) of the variance of fatigue. Extraversion and optimism correlated negatively with fatigue.

Table 4. Correlations for personality dimensions and total scores of FSS-9 and FSS-7.

Variables		N	E	O	A	C	Op
FSS-9	r	.531**	-.474*	-.186	-.237	-.203	-.424*
	p	.008	.019	.385	.265	.342	.039
	CI	[.160, .759]	[-.747,-.033]	[-.552, .193]	[-.552,.130]	[-.599,.212]	[-.713,-.066]
FSS-7	r	.574**	-.515**	-.100	-.235	-.224	-.469*
	p	.003	.010	.641	.270	.292	.021
	CI	[.245,.767]	[-.762,-.082]	[-.508,.291]	[-.541,.133]	[-.595,.156]	[-.732,-.149]

Note: CI = 95% bootstrapped confidence intervals; FSS = Fatigue Severity Scale; NEO-FFI = NEO Five-Factor Inventory, Neuroticism (N); Extraversion (E); Openness to experience (O); Agreeableness (A); Conscientiousness (C) and Optimism (Op); Fatigue Severity Scale FSS

In addition to the correlations in Table 4 the correlation with age and gender was calculated, but this did not affect the results.

Discussion

This study aimed to investigate the relationship between poststroke fatigue in daily life and personality traits. Momentary general fatigue was negatively and significantly associated with extraversion and with optimism. That is, individuals who scored high on extraversion and optimism reported a lower score on momentary general fatigue. No associations were found between momentary general fatigue and the other personality traits. With respect to momentary mental fatigue, a significant negative association was found with extraversion, conscientiousness and optimism. Individuals who scored higher on extraversion, conscientiousness, and optimism reported a lower score on momentary mental fatigue. No significant associations were found between momentary physical fatigue and personality traits. A moderate positive correlation was found between retrospective fatigue reports and neuroticism, and the retrospective fatigue scale correlated negatively with extraversion and optimism. Individuals who scored high on this scale reported higher scores on neuroticism and lower scores on extraversion and optimism.

Surprisingly, no association was found between momentary general fatigue and momentary physical or mental fatigue and neuroticism. When fatigue was measured with retrospective fatigue questionnaires, we found a moderate positive correlation with neuroticism. An explanation for these results may lie in the way that fatigue was measured. Symptoms that are collected with the ESM method may be less prone to symptom amplifying thoughts because this method better captures moment-to-moment fluctuations and may therefore reveal experiential knowledge provided by the experiencing self rather than by the remembering self (Van Den Bergh & Walentynowicz, 2016; Lenaert et al. 2020). That is, individuals may engage less in cognitive appraisals of their symptoms when asked to report them in the current moment relative to retrospective reports (Heron & Smyth, 2010). When it comes to retrospective questions, individuals have to reconstruct symptom experience over a longer period of time which is based on autobiographical memory, which may also leave more room for cognitive evaluations or symptom amplifications (Robinson & Clore, 2002).

Individuals with stroke who are more extravert experience less momentary fatigue and tend to report less fatigue retrospectively. Michielsen et al. (2007) also found a negative association between extraversion and fatigue in individuals with breast cancer. They argued that two mediating links may exist between extraversion and fatigue: the amount of daily activity and being cheerful. The NEO-FFI extraversion factor consists of several facets, one of which refers to undertaking various daily activities. Thus, Michielsen et al. argued that the relationship between extraversion and fatigue might be mediated by the amount and variety of daily activities. Further, positive emotions are another facet of extraversion and are strongly associated with happiness and well-being, which may also explain why highly extravert individuals experience less fatigue or less impact thereof on daily life.

Relatedly, our results are in line with other studies in individuals with cancer that reported a negative association between fatigue and optimism (Allison et al., 2003; Chambers et al., 2012; Person et al., 2020). That is, optimistic patients, appear to experience less fatigue. Optimism or dispositional optimism has been associated with psychological well-being, physical health, health behaviour, and more self-esteem (Solberg Nes & Segerstrom, 2006). Research suggests that optimistic persons tend to pay more attention to positive events and also engage in more effective coping or problem solving (Scheier & Carver, 2018).

Our findings may have implications for the diagnostic and treatment phase of patients with poststroke fatigue. In the diagnostic phase of poststroke fatigue, professionals should consider how personality characteristics may influence symptom reporting. Although it may not be feasible to always administer a comprehensive personality interview when assessing fatigue severity, heightened awareness of the potential association between symptom reporting and personality may enable clinicians to take this into account in the diagnostic process.

Concerning treatment, individuals high in optimism may have more effective coping strategies in dealing with complaints and problems during rehabilitation, whereas less optimistic persons may lack these strategies (Mavaddat et al., 2017). Solberg Nes and Segerstrom (2006) found that dispositional optimism was positively correlated with problem-focused and emotion approached coping. In dealing with challenging situations, optimistic persons frequently use strategies such as active and task-oriented coping and positive reinterpretation, acceptance or cognitive restructuring. From this perspective, it is possible that optimistic persons use more effective coping strategies when confronted with a life-changing event such as stroke or disease-specific stressors such as fatigue.

Although speculative because these results do not allow causal inferences on the relation between optimism and fatigue, it would be interesting to investigate whether less optimistic individuals would benefit from certain add-on treatments. For instance, research suggests that dispositional optimism (Malouff & Schutte, 2017) and optimistic attributional styles (Seligman, 2011) can to a certain degree be learned. To the extent that a causal relationship between optimism and fatigue exists, learned optimism may also contribute to improvements in fatigue.

Furthermore, extraversion is known to have different facets, such as sociability and energy (Carver & Connor-Smith, 2010). Persons who score high on these facets engage more in activities with others and may engage in more different activities (Carver & Connor-Smith, 2010). This information can be used in the treatment phase of fatigue where more extravert individuals may for instance, adhere better to group treatment in which they carry out assignments together with others. Further research is needed to investigate this relationship. For instance, it is unclear whether individuals low in extraversion are also at greater risk for poor rehabilitation outcomes related to fatigue.

A significant strength of this study is the measurement of poststroke fatigue using two measurement methods, namely ESM and retrospective questionnaires, thereby revealing that personality is related to poststroke fatigue and that this relation may differ according to the measurement method used. Further, our study is the first to explore the connection between personality and momentary fatigue after stroke. This demonstrates that relatively stable person characteristics are related to diurnal variations in fatigue experience. When rehabilitation treatment seems to stagnate or little improvement in fatigue is achieved, it may be appropriate to take into account person characteristics and their relationship to momentary fatigue.

Limitations include the relatively small sample size. This study was largely exploratory in nature, and therefore, replication studies with larger samples are needed. Further, the sampling period was restricted to six days. Future research could assess the relationship between personality traits and poststroke fatigue over longer periods. For example, it would be interesting to know how poststroke fatigue, measured with ESM, evolves over time or during the rehabilitation period and how personality traits are related to it (Wijenberg et al., 2019). However, keeping track of the complaints over longer time periods may present a burden on patients. An alternative could be to reduce the measurement days and measurement moments. A period of six weeks, including 3 ESM assessment days per week with 8 beeps per day, has been used with good reponse and compliance rates (Kramer et al., 2014; van Knippenberg et al., 2018). Third, personality traits were investigated after stroke and were assessed retrospectively by participants. It is possible that a cerebrovascular accident is causally involved in both personality change and higher levels of fatigue and their interrelationship. In such case, associations between poststroke personality and fatigue could merely reflect that the long-term consequences of stroke can be associated or may reinforce each other. Ideally, personality (change) is measured before and after stroke, but this is difficult to achieve in practice. To gain more insight into pre-stroke personality, a questionnaire completed by a close relative could be used (Wijenberg et al., 2019). A fourth potential limitation of this study refers to the relatively young age of our sample and the study requirement of owning a smartphone and being cognitively capable of using the ESM application. The average age of our sample was 55.5 years (SD = 7.6, range: 34 - 68). Therefore, some caution is warranted when generalizing the current findings to older or more cognitively impaired individuals with stroke.

To conclude, this exploratory study extends our knowledge of the associations between personality traits and poststroke fatigue. We showed that poststroke fatigue is associated with some personality traits, but this also depends on the way poststroke fatigue is measured. These results can play an important role in the diagnostic and treatment phase of poststroke fatigue. We recommend, when investigating the relationship between poststroke fatigue and person characteristics such as personality, to include not only retrospective but also momentary measures of fatigue as they may reflect different information about the perception of fatigue.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.