Patients (≥18 years) in the chronic phase (≥1 year) after treatment at our hospital for aSAH between January 2012 and March 2018, and permanently living in the South-Eastern Norway Regional Health Authority were eligible. A clinical neuropsychologist conducted the recruitment by telephone interview. All patients with a mean FSS score ≥4 were invited for assessment at the hospital. We excluded patients unable to conduct the different assessments, handle the instruments used for evaluation, or those who were unable to give consent. We also excluded those that had undergone brain surgery <12 months prior to inclusion, patients with brain tumor, cerebral arteriovenous malformation, inadequately treated hydrocephalus, epilepsy, cerebral paresis, and neurodegenerative disease. Pregnancy also prohibited participation, as did active drug abuse, severe blood test deviations from normal and an electrocardiogram with prolonged QTc time >480 ms.

Due to possible interactions with (-)-OSU6162, we further excluded patients using metabolic enzyme inhibitors and inducers and those using drugs with a narrow therapeutic window or requiring individual dose adjustments. The following Anatomical Therapeutic Chemical (ATC) Classification System categories were not allowed: N06B A+X, N07A, N06A G+X, N05A, N03A, J04A, J01D H, L04A D, B01A A, C01A A, H01B A, and N04B D.

Statistical analysis was performed with IBM SPSS version 25 for Windows (Armonk, NY: IBM Corp). Continuous variables were presented by mean and standard deviation, and independent samples t-test was used to compare differences between groups. Continuous variables which were not normally distributed were presented with median and range, and the Mann–Whitney U test was used for differences between groups. Categorical variables were presented as frequencies or percentages, and we used the Chi-Square test to compare differences between groups. The itemized analysis was conducted by comparing the mean FSS and mean MFS item scores against the mean FSS score ± 2 SEM and mean MFS item score ± 2 SEM for the whole group (N = 96). FSS and MFS item scores defined as prominent features of post-aSAH fatigue also corresponded to results of skewness (i.e., highest negative values). Bivariate and partial Pearson correlation coefficient was used to explore FSS and MFS with continuous variables and adjust for covariates. Statistical significance was set at 0.05 (two-sided).

A total of 749 patients were admitted for aSAH between January 2012 and March 2018, of whom 677 received active treatment, 430 patients were eligible for a telephone interview, 368 completed prescreening by telephone and 96 patients completed the full assessment at the hospital (see Figure 1). The time from hemorrhage to inclusion was median 25 months (range 12–83 months). The characteristics of the 96 included patients are displayed in Table 1. Whereas the 96 included and the 272 excluded patients did not differ with respect to gender [X(1)2 = 0.02, p = 0.886], aneurysm location [X(1)2 = 0.38, p = 0.540] or mode of aneurysm repair [X(1)2 = <0.01, p = 0.969], the excluded patients were significantly older at the time of hemorrhage [t₍₁₎ = 3.58, p = <0.001] and in better clinical condition (HH 1–3 vs. HH 4–5) prior to aneurysm repair as compared to the included patients [X(1)2 = 4.29, p = 0.038].

For all 96 patients, the mean FSS score was 6.0 ± 0.8, and the MFS sum score was 18.1 ± 5.6. Scores on the two fatigue measures were closely connected, where higher scores on FSS were significantly related to higher scores on MFS [r₍₉₄₎ = 0.47, p = <0.001].

Figures 2 and 3 show which of the nine FSS items and the 14 MFS items that stood out as most and less prominent by displaying the mean scores for each item for the entire group and those with mood disorders against the mean FFS score ± 2 SEM and the mean MFS item score ± 2 SEM.

On the FSS questionnaire (Figure 2), the two items that were scored highest by the entire group was “My motivation is lower when I am fatigued” (item 1: 6.5 ± 1.0) and “Fatigue is among my three most disabling symptoms” (item 8: 6.6 ± 0.7). The items “Exercise brings on my fatigue” (item 2: 4.7 ± 2.0) and “Fatigue causes frequent problems for me” (item 5: 5.6 ± 1.4) had the lowest scores; however, within these two items we found the largest differences between all patients and the subgroup with mood disorders. Item 3 “I am easily fatigued” fell just below the ± 2SEM of mean FSS for the entire group, but scored within this range in those with mood disorders. All other items on FFS were above or within the range of cut-off (± 2 SEM).

On the MFS questionnaire (Figure 3), the items “Mental fatigue” (item 3: 1.9 ± 0.5) and “Sensitivity to stress” (item 8: 1.9 ± 0.9) stood out with the highest scores. Item 11 “sensitivity to light” (0.9 ± 0.7) and item 13 “decreased sleep” (0.8 ± 0.8) scored lowest. Item 6 “Memory problems”, item 9 “Emotional instability”, and item 10 ”Irritability” scored below the range of cut-off (± 2 SEM) for all 96 patients but scored within or above the range of cut-off in the subgroup of patients with mood disorders. All other items on MFS were above or within the range of cut-off (± 2 SEM).

The proportion of patients within the different standardized categories for BDI-II and BAI are listed in Table 2. Clinical depression and clinical anxiety were reported by 34.4% and 18.8%, respectively.

There was a positive correlation between mean FSS scores and depressive symptoms (BDI-II) scores with and without adjustment for anxiety symptoms (BAI) score [r₍₉₄₎ = 0.42, p = <0.001; r_adj.(93) = 0.34, p = <0.001]. In contrast, the relationship between mean FSS scores and anxiety symptoms (BAI) scores disappeared when adjusted for depressive symptoms (BDI-II) [r₍₉₄₎ = 0.27, p = 0.007; r_adj.(93) =−0.02, p = 0.833]. A positive correlation was found between MFS sum scores and depressive symptoms (BDI-II) scores, also when adjusted for anxiety [r₍₉₄₎ = 0.52, p = <0.001; r_adj.(93) = 0.30, p = 0.003]. Although still reaching significance, the correlation between MFS sum scores and anxiety symptoms (BAI) was weakened when adjusted for depressive symptoms (BDI-II) [r₍₉₄₎ = 0.49, p = <0.001; r_adj.(93) = 0.21, p = 0.043].

Symptoms of depression and anxiety (BDI-II and BAI scores) within demographics, prior history, and aSAH characteristics are presented in Table 3. Patients with prior history of depression had both higher BDI-II scores (depressive symptoms) and higher BAI scores (anxiety symptoms) than those without a prior history of depression. In contrast, patients with prior history of anxiety had higher BAI scores (anxiety symptoms), but not higher BDI-II scores (depressive symptoms) than those with no prior history of anxiety. Except for a trend towards higher BDI-II scores (depressive symptoms) among patients with endovascular as opposed to surgical aneurysm repair (17.8 + 8.6 vs. 14.4 + 8.7, p = 0.058), demographical data and aSAH characteristics were not related to BDI-II and BAI scores. Mood disorders (depression and anxiety) were more common in patients undergoing endovascular than surgical aneurysm repair [47.2% vs. 23.3%, p = 0.019].

Fatigue scores vs. demographics, prior history, and aSAH characteristics are presented in Table 4. Patients with low/intermediate education level had higher mean FSS scores than those with high education level (6.3 ± 0.7 vs. 5.7 ± 0.8, p = 0.001) and patients with endovascular aneurysm repair had higher mean FSS scores than those with surgical aneurysm repair (6.2 ± 0.7 vs. 5.7 ± 0.8, p = 0.002). There were no significant relationships between MFS sum scores and any of the background variables.

There was no significant relationship between fatigue and neurological impairment as scored with NIHSS [FSS: r₍₉₄₎ = 0.06, p = 0.583; MFS: r₍₉₄₎ = 0.17, p = 0.094]. There was also no significant relationship between fatigue and clinical outcome as scored with mRS [FSS: r₍₉₄₎ = −0.06, p = 0.578; MFS: r₍₉₄₎ = 0.17, p = 0.093] where all mRS 0 patients [n = 5] scored ≥4 on FSS and 60.0% [n = 3] scored ≥10.5 on MFS.

Weight gain was significantly associated with mean FSS scores [r₍₉₄₎ = 0.25, p = 0.016], but not with MFS sum scores [r₍₉₄₎ = −0.16, p = 0.128]. Further, weight gain was associated with depressive symptoms (BDI-II) [r₍₉₄₎ = 0.27, p = 0.009], but did not relate to anxiety symptoms (BAI) [r₍₉₄₎ = 0.06, p = 0.572]. When adjusted for depressive symptoms, mean FSS scores was no longer associated with weight gain [r_adj.(93) = 0.15, p = 0.140].

The neuropsychological test performances of the 96 patients are shown in Figure 4 (see also Supplementary Material: Neuropsychological test performance). As illustrated, the percentages of deficits were low and ranged from 3.9% to 8.7% within the six cognitive domains. The highest percentages of deficits were found for the domains “sensomotor function” [8.7%], “verbal memory” [7.6%], and “executive function” [6.9%].

Digit Span Forward (WAIS-IV) was negatively correlated to mean FSS scores [r₍₉₄₎ = −0.21, p = 0.040, corrected: p = 0.960] and CVLT-II recognition Hits was negatively correlated with MFS sum scores [r₍₉₄₎ = −0.22, p = 0.030, corrected: p = 0.720]. These associations were still significant when adjusted for mood disorders [r_adj.(93) = −0.25, p = 0.013, corrected: p = 0.312; r_adj.(93) = −0.26, p = 0.011, corrected: p = 0.264, respectively]. Grooved Pegboard dominant was negatively correlated with MFS sum scores [r₍₉₄₎ = −0.20, p = 0.046, corrected: p = 1.0], but not when adjusted for mood disorders [r_adj.(93) = −0.19, p = 0.062]. However, none of the correlations remained statistically significant after Bonferroni correction. Hence, both fatigue scores were unrelated to the 24 neuropsychological test performance scores.

Table 2 presents the results for the SF-36 subscales. Approximately half of the patients scored low (experienced problems) on the subscales role-physical, vitality, and social functioning.

We found significant negative correlations between both mean FSS and MFS sum scores and all SF-36 subscales; i.e., more fatigue correlated with poorer HRQoL (lower scores). All correlations reached p = ≤0.001 significance level except for the associations between fatigue scores and General health subscale which reached a p = <0.05 significance level [FSS: r₍₉₄₎ = −0.32, p = 0.002; MFS: r₍₉₄₎ = −0.20, p = 0.044]. After adjusting for mood disorders, mean FSS and MFS sum scores were still negatively correlated with all SF-36 subscales [p = <0.05] except for the association between MFS sum scores and General health subscale [r_adj.(93) = −0.07, p = 0.483].

Among the 78 patients that were employed at the time of hemorrhage; 43 [55.1%] had not returned to work (no RTW), 27 [34.6%] had partial RTW, and eight [10.3%] had full RTW. There was a significant difference between the mean FSS score for the different RTW categories (F_(2,75) = 4.17, p = 0.019, Figure 5A). The mean FSS score was significantly lower in patients with full RTW [5.40 ± 0.60] as compared to no RTW [6.17 ± 0.69, p = 0.005], but not when compared to partial RTW [5.84 ± 0.87, p = 0.085]. No significant difference in mean FSS score was seen between partial RTW and no RTW [p = 0.196]. Figure 5B shows the MFS sum score for the three RTW categories, which were not significantly different [F_(2,75) = 1.50, p = 0.231]. There was further no significant difference in depressive symptoms (BDI-II) scores [F_(2,75) = 1.65, p = 0.199] or anxiety symptoms (BAI) scores [F_(2,75) = 2.93, p = 0.060] for the different RTW categories.

Our itemized analysis favored high scores for questionnaire items that mostly are linked to a mental type of fatigue, whereas those linked to physical activity scored relatively low. Patients reported a disproportionately large drainage of mental energy after executing cognitive tasks or after engaging in a conversation with several people (MFS item 3). Motivation levels were low when fatigued (FSS item 1) and there was a reduced ability to cope with stress and manage tasks under time pressure (MFS item 8). These are typical features of mental fatigue, described as a dynamic process with fluctuation of mental energy levels. Buunk et al. (2018) found a frequency of mental fatigue of 48.4% and of physical fatigue of 38.5% in patients with aneurysmal and non-aneurysmal SAH, which indicates a stronger component of physical fatigue than our results suggest. The instrument used for measuring fatigue was, however, different as Buunk et al. (2018) used The Dutch Multifactor Fatigue Scale. Sörbo et al. (2019) investigated post-aSAH fatigue with the MFS questionnaire only and found 57% of patients scoring ≥10.5 points. They used no other questionnaires and their evaluation of physical fatigue was hence not beyond the means of the present study.

A relationship between pre- and post-stroke fatigue has been reported both in the acute (Lerdal et al., 2011) and chronic phase (Choi-Kwon et al., 2005; Lerdal et al., 2012). We did not find a significant difference between scores of fatigue in patients with or without a prior history of fatigue. To the best of our knowledge, no data regarding the relationship of pre- and-post-aSAH fatigue have previously been published. We interpret this result, in line with the other findings in the present study, to further support the notion that post-aSAH fatigue is primarily a result of pathological mechanisms caused by the hemorrhage itself.

Although the prevalence of mood disorders after aSAH is high (Al-Khindi et al., 2010; Rinkel and Algra, 2011; Vetkas et al., 2013; Tang et al., 2020), few studies have examined the relationship between mood disorders and post-aSAH fatigue. Passier et al. (2011a) found higher fatigue scores 1 year after the ictus in patients scoring high on anxiety and depression than in those without such complaints 3 months after aSAH. Our results are in line with this; i.e., post-aSAH fatigue was strongly associated with depression. Furthermore, post-aSAH fatigue was to some degree associated with anxiety, but to a lesser extent than depression.

Even though post-aSAH fatigue and mood disorders were strongly associated, the overlap was incomplete since a majority of our patients did not have clinical depression or anxiety (65.6% and 81.2%, respectively). Stroke studies have demonstrated that fatigue may occur in the absence of depression (Ingles et al., 1999; van der Werf et al., 2001; Choi-Kwon et al., 2005). Our results suggest that this is also the case for fatigue and mood disorders after aSAH. Fatigue and mood disorders can exist independently, supporting the more commonly accepted notion that these two are distinct entities. Our itemized analysis of the fatigue questionnaires adds to the existing literature that the profile of fatigue characteristics is reported differently in those with post-aSAH fatigue and those with mood disorders. This latter subgroup reported more often that exercise brings on fatigue (FSS item 2), that fatigue causes frequent problems (FSS item 5), and that fatigue causes frequently problems with work, family, and social life (FSS item 9). Furthermore, those with mood disorders reported higher scores for sensitivity to stress (MFS item 8), emotional instability (MFS item 9), irritability (MFS item 10), and sensitivity to noise (MFS item 12). This subgroup may therefore need special attention during rehabilitation.

The temporal relationship between fatigue and mood disorders is not well understood. It is uncertain whether mood disorders after hemorrhage are influenced by personal factors, the aSAH itself, or by its consequences. We found no relationship between mood disorders and aSAH characteristics, which might suggest that mood disorders after aSAH are less likely to be a consequence of direct organic brain injury. However, patients with prior history of mood disorders, and especially depression, had higher scores on BDI-II and BAI than patients without this prior history. A suboptimal coping style could possibly serve as a predisposition for mood disturbance after aSAH (Noble et al., 2008) and therefore indirectly be related to post-aSAH fatigue.

Previous studies have failed to show significant relationships between post-aSAH fatigue and age, gender, education level, aneurysm localization, and treatment modality (Passier et al., 2011a; Western et al., 2020). We presently found higher mean FSS scores but similar MFS sum scores among those with low/intermediate as compared to high education level. Selection bias might be of importance as our aSAH population was on average highly educated, whereas the population in Passier et al. (2011a) had primarily low education. The higher mean FSS scores in our patients with endovascular aneurysm repair could be due to the larger fraction of patients with mood disorders in that group.

Sörbo et al. (2019) demonstrated a correlation between MFS score and functional outcome after aSAH as expressed with GOSE, where none of their fully recovered patients (GOSE 8) had an MFS sum score suggestive of mental fatigue (≥10.5). Presently, all five mRS 0 patients scored ≥4 on FSS and three of them scored ≥10.5 on MFS. The mRS may be less sensitive to certain functional impairments as compared to the GOSE. The mRS was presently scored by the clinician during the control visit. A score of mRS 0 may indicate that significant fatigue may be overlooked in a regular interview, especially in patients that have no neurological impairments. Due to the exclusion criteria of our study, we included patients that had very few neurological impairments, and hence one would not expect to find a clear relation between neurological status and fatigue.

Depressed individuals are at higher risk for developing obesity than non-depressed individuals (Blaine, 2008). This can explain that weight gain in our aSAH patients was related to depressive symptoms and not to fatigue. Weight gain was hence no surrogate marker for physical fatigue. Even though FSS item 2 “Exercise brings on my fatigue” was scored higher in those with mood disorders, it was the least prominent FSS item also in that subgroup.

Numerous studies have documented cognitive sequelae after aSAH (Al-Khindi et al., 2010; Rinkel and Algra, 2011; Nordenmark et al., 2019b; Burke et al., 2020; Nussbaum et al., 2020). Despite increasing interest in using neuropsychological tests in fatigue research, a lack of correlation between fatigue and neuropsychological performance has been observed across a wide variety of clinical samples (DeLuca, 2005). This concurs with the present findings where few of our patients had cognitive deficits despite all of them suffering from fatigue. According to DeLuca (2005), the most consistent finding is for subjective fatigue to be more closely related to depression than with objective cognitive performance. This is also in line with our results.

Passier et al. (2011a) demonstrated that aSAH patients with cognitive impairments reported a higher level of fatigue than those without cognitive impairment. In their subgroup of patients with neither physical nor cognitive impairment, however, passive coping style and emotional problems were considered important predictors of fatigue. This could possibly also be so in the present study as our patient cohort of 96 patients closely resembled their subgroup of patients without physical and cognitive deficits. Assessment of coping style among our patients could hence have provided valuable information regarding fatigue.

It is well-documented that reduced HRQoL is a common sequel after aSAH (Visser-Meily et al., 2009; Al-Khindi et al., 2010; Rinkel and Algra, 2011; Wong et al., 2011; Czapiga et al., 2013; Passier et al., 2013; Taufique et al., 2016). Visser-Meily et al. (2009) found post-aSAH fatigue to be strongly related to a decreased HRQoL. Our findings, where both high mean FSS scores and MFS sum scores were strongly associated with abnormal findings in all subscales of SF-36, concur with that. Even after adjusting for mood disorders, known to be related to reduced HRQoL (Visser-Meily et al., 2009), most of the subscales on SF-36 presently correlated with the post-aSAH fatigue scores.

Limitation in physical and social activities and low vitality were presently reported as the most reduced aspects of HRQoL. Despite their good neurological function, half of our patients still reported physical limitation as the most affected impairment. This was mainly due to their fatigue and not because they actually were physically impaired. Czapiga et al. (2013) also found this subscale to be most impaired in an aSAH population so that this could possibly reflect a physical component of fatigue. A limitation of social activities was also reported as one of the most affected aspects of quality of life. Drainage of mental energy and stress hyper-sensitivity poses a challenge for engagement in social activities. Social withdrawal may therefore be interpreted as a coping mechanism to reduce fatigue.

Despite high functional independence and low frequency of cognitive deficits, we found higher mean FSS scores to be related to a lower rate of RTW. This may be due to a partial overlap in measurement as FSS quantifies the impact of fatigue on daily life. Measuring fatigue with the MFS could not reproduce that clear relationship to RTW, but may perhaps have done so in a larger sample. Mood disorders could not explain our low rate of 10.3% of RTW. Several studies have reported that aSAH survivors have a surprisingly low rate of RTW although being physically capable of working (Hop et al., 2001; Wermer et al., 2007; Passier et al., 2011b; Wallmark et al., 2016; Nordenmark et al., 2019a). Even though the low rate of RTW after aSAH probably is a multifactorial problem, our study demonstrates that fatigue is an important factor to consider when planning occupational therapy. The consequence of the low rate of RTW is not only of economical and psychosocial importance for the relatively young aSAH population but also for society.

Some have questioned if a so-called good outcome justifies the assumption that aSAH patients have no relevant neurobehavioral impairments (Hütter et al., 1999). Our findings support the notion of an invisible dysfunction after aSAH. All our patients were good outcome (mRS 0–2) where majority had no signs of cognitive deficit, nevertheless, they experienced a debilitating and long-lasting fatigue, sometimes in conjunction with emotional problems, with a major impact on perceived quality of life and ability to return to work. This apparent discrepancy is of importance in clinical practice. It is crucial that clinicians early in the recovery process can identify and acknowledge that fatigue is distinct from normal exhaustion and may be part of a long-lasting illness trajectory with a huge impact on social and occupational engagement. The aSAH survivor and their families are in need of information about the prevalence and typical characteristics of post-aSAH fatigue in order to have a better foundation for realistic expectations regarding recovery and coping. Acknowledging post-aSAH fatigue as a health problem can facilitate coping.

Since our results support the notion of fatigue and depression as distinct entities, it appears sensible to treat the latter with antidepressants. A relief of depressive symptoms has the potential to reduce symptom severity of fatigue although not being a cure of fatigue per se. Fatigue as the predominant contributor of the inability of RTW will, however, not be affected by antidepressant treatment. Since physical fatigue and impairment is not a dominant problem, principles of stroke rehabilitation with a stronger focus on physical training may not be helpful or even counterproductive in aSAH patients (Johansson and Rönnbäck, 2014).

Our findings also highlight the need for more research. The present study emphasizes the need for more knowledge on the typical features of fatigue after aSAH. A better understanding of which fatigue features to focus on and how to measure these aspects in a standardized manner could generate empirical evidence about the ossible underlying mechanism of fatigue and will therefore endorse the development of evidence-based treatment for fatigue. To this date, there is still insufficient evidence to support the use of any intervention to treat or prevent fatigue after stroke (Wu et al., 2015).

Our results have to be interpreted in light of the strict selection of participants based on the inclusion and exclusion criteria of the RCT they were recruited into. This resulted in all participants being good outcome (mRS 0–2) aSAH survivors without significant neurological or cognitive deficits. Furthermore, participation in our RCT, from where the present data were extracted, was comprehensive, many patients would have to travel over long distances to participate, and many patients, therefore, declined to participate. The RCT participants may thus have unknown common features affecting the profiles presently investigated. Nevertheless, since there were no significant differences between included and excluded patients with regard to gender, aneurysm location, and mode of aneurysm repair, and as we included patients over the entire specter of aSAH severity (even more in poor grade than among those excluded), the present findings should be representative for all patients with post-aSAH fatigue. It appears paradoxical that the deadliest type of intracerebral hemorrhage results in more neurologically intact survivors than ischemic stroke; it underlines, however, that the post-aSAH fatigue group is diverse from the post-stroke group that often struggles with neurological deficits. Selecting patients with post-aSAH fatigue and significant neurological impairment may have produced a different nature of fatigue in our itemized analysis. On the other hand, including only good outcome patients pinpoints the detrimental effects of fatigue after aSAH.

There are no validated instruments to assess fatigue after aSAH. Although the mental features of fatigue are not properly measured with FSS, it is still the most frequently used questionnaire for evaluating fatigue in stroke studies due to its high internal consistency (Nadarajah et al., 2017). The MFS is a new questionnaire for assessing mental fatigue after mild TBI, and its psychometric properties are not extensively studied in the aSAH population. On the other hand, there are strong similarities to the clinical picture in the chronic phase after mild TBI and aSAH, suggesting that the MFS questionnaire may also be a useful tool in the evaluation of post-aSAH fatigue. The MFS incorporates items addressing cognitive complaints like “Concentration difficulties” (item 5), “Memory problems” (item 6), and “Slowness of thinking” (item 7) which are not directly linked to fatigue. The score on these items will hence increase the MFS sum score, without a higher score necessarily reflecting more fatigue. It is noteworthy that our patients scored all of these three items higher than 1.0 on average even though the majority of their neuropsychological test performances were within the normal range. The individual perception of cognitive problems may hence also be an aspect of fatigue. Future studies of fatigue after aSAH should therefore employ and validate the use of questionnaires that address the multidimensionality of post-aSAH fatigue.

Also, the BDI-II and to some extent the BAI contain items that could be experienced by patients with post-aSAH fatigue but not necessarily the symptoms for which the measures were developed. Although we attempted to correct for this overlap by setting a conservative cut-off for the BDI-II and BAI, the potential for symptom overlap still exists.

Good outcome patients with post-aSAH fatigue experienced their fatigue as being among the three most disabling symptoms and when characterizing their fatigue they emphasized the questionnaire items “low motivation,” “mental fatigue,” and “sensitivity to stress”. Fatigue due to exercise was the least bothersome aspect of fatigue and weight gain was associated with depressive symptoms rather than the severity of fatigue. Although there is a strong association between post-aSAH fatigue and mood disorders, especially depression, the overlap is incomplete and our findings support the notion that these symptoms are distinct entities. Post-aSAH fatigue often exists without significant neurological or cognitive impairments but relates significantly to reduced HRQoL and contributes to a low rate of return to work.