Associations between psychological factors and health-related quality of life and global quality of life in patients with ALS: a systematic review

Objective To systematically identify and appraise evidence on associations between psychological factors (moods, beliefs, personality) and Health-related QoL (HRQoL) and/or global QoL in patients with Amyotrophic Lateral Sclerosis (ALS). Methods A systematic review was conducted in several online databases (PsycINFO, EMBASE, PubMed and CINAHL) up to October 2015. Articles were included if they reported associations between psychological factors (moods, beliefs and personality) and HRQoL and/or global QoL in an ALS population. The search was limited to empirical studies, published in English, which provided quantitative data. The methodological quality of the included articles was assessed. Results In total, 22 studies were included. Mood was investigated in 14 studies, beliefs in 11 studies and personality in one study. Fifteen different psychological factors were extracted and assessed using 24 different measures. Twelve different QoL measures were used in the selected studies, subdivided into seven different HRQoL measures and five different global QoL measures. Higher levels of anxiety and depression appeared to be related to a poorer HRQoL, whereas a higher level of religiosity seemed to be associated with better global QoL. No conclusive associations were found for confusion-bewilderment (mood), spirituality, mindfulness, coping styles, hopelessness, perception of burden, cognitive appraisal (beliefs), neuroticism, extraversion, openness, agreeableness and conscientiousness (personality), due to insufficient or inconsistent evidence. Religiosity and spirituality appeared to become more positively associated over time. Conclusions Our results suggest that higher levels of anxiety and depression are related to a poorer HRQoL, whereas higher levels of religiosity appeared to be related to better global QoL. Associations might change during the disease course. This review supports the importance of psychological factors with regard to ALS care. Further research is needed to supplement the available evidence and to investigate how psychological factors can be modified to improve QoL. Review registration number PROSPERO 2015:CRD42015027303 Electronic supplementary material The online version of this article (doi:10.1186/s12955-016-0507-6) contains supplementary material, which is available to authorized users.


Background
Amyotrophic Lateral Sclerosis (ALS) is a fatal, progressive, neurodegenerative disorder affecting motor neurons in the spinal cord, brainstem and motor cortex. Patients suffer progressive wasting and weakness of limb, bulbar and respiratory muscles, leading to inability to speak and swallow, respiratory failure and complete paralysis [1,2]. Currently, there is increasing awareness that ALS is also associated with non-motor findings, including behavioral and cognitive deficits [3,4]. Patients eventually die due to respiratory failure within three to five years after symptom onset [1]. The incidence of ALS shows little variation in Western countries, ranging from 1.5 to 2.7 per 100,000 person-years, [5] with an estimated lifetime risk of 1 in 400 [6]. To date, no curative treatment is available. Therefore, optimal treatment is based on symptom management and optimizing Quality of Life (QoL).
There is, as yet, no agreed-upon definition of QoL. The World Health Organization (WHO) defines QoL as 'a broad ranging concept affected in a complex way by the person's physical health, psychological state, level of independence, social relationships and their relationship to salient features of their environment' [7]. Burns et al. [8] have suggested that a distinction can be made between Health-related QoL (HRQoL) and global QoL in patients with ALS. Health-related QoL (HRQoL) is more narrowly defined than global QoL and seeks to address those aspects of self-perceived well-being that are related to or affected by the presence of disease or treatment [8]. Assessment of HRQoL typically includes physical, psychological and social domains. Each domain may include measures that assess the patient's perception of symptoms, ability to function and disability [9]. Global QoL reflects overall QoL as judged by the patient and takes into account other, non-medical concepts, such as family, support system and friends [8]. Assessing global QoL generally provides a broader picture of the impact of disease on an individual's life [9].
HRQoL declines during the course of the disease [10]. This is expected as HRQoL instruments are heavily weighted toward physical function, and thus inevitably decline over time as patients with ALS lose their abilities. In contrast, there is growing evidence that global QoL seems to remain at a stable level, even in patients with advanced ALS. Psychological processes like coping, reframing expectations and spiritual practice might contribute to a change in internal standards and values of QoL, ultimately resulting in unexpectedly high QoL, even in later disease stages [11,12]. QoL in patients with ALS seems to be determined more by psychological, existential and support factors than by physical health [13][14][15][16], implying that a broad range of factors is involved in adjusting to illness. Psychological factors in patients with ALS may be modifiable targets for interventions to improve QoL.
The impact of psychological factors such as neuroticism, coping, cognitive appraisals and mood on QoL has already been demonstrated in other chronic diseases [17,18] and in other progressive neurological illnesses, such as Huntington's disease, Parkinson's disease and multiple sclerosis [19]. Differences have, however, been reported between these progressive neurological illnesses and ALS concerning the contribution of psychological factors to QoL [19], suggesting that the rapidly progressive disabling process of ALS requires a different psychological adaptation process.
Over the last decade, interest has grown in the relationships between psychological factors and QoL in patients with ALS and to date, three narrative reviews on this subject have appeared [20][21][22]. The authors summarized associations between QoL and depression [20,21], anxiety [20], spiritual and existential issues [20][21][22], sense of burden [22] and hope/hopelessness [20,22] in patients with ALS, but as they did not quantify or appraise them, the relationships remain unclear.
The present study aims to collect and appraise the available evidence on the associations between psychological factors and HRQoL and/or global QoL. Understanding the relationships between QoL and psychological factors and the contribution of these factors to either HRQoL or global QoL might help health professionals to develop adequate interventions in order to optimize QoL in patients with ALS.
another author (JV) was consulted for a final judgment. The same procedure was followed for final in-or exclusion after reading full text articles. The reference sections of retrieved articles were searched to identify further studies suitable for inclusion.

Quality assessment
After the study selection, methodological quality was assessed independently by two researchers (AvG, EK) according to an 8-point checklist, resulting in a score that ranged from lowest quality (1) to highest quality (8) [17]. The level of agreement between the researchers' ratings was established using the Intraclass Correlation Coefficient (ICC).

Eligibility criteria and operationalization of concepts
The current review is restricted to empirical studies which provided quantitative data, thus excluding qualitative studies, reviews and case reports. Only studies of patients with ALS or providing separate data from patients with ALS were included, in which standardized measures were used to assess direct relationships between psychological factors (determinant) and a total QoL construct (outcome). Studies using a total score for HRQoL and global QoL, or a mental or physical component score of the HRQOL and / or a single-item score representing global QoL, were included. Thus studies describing associations between psychological factors and one subscale of a QoL measure were not taken into account. Furthermore, the review was limited to studies written in the English language that were published in peer-reviewed journals.
Psychological factors are part of the contextual factors (personal and environmental factors) defined by the International Classification of Disability, Functioning and Health (ICF) [24]. Psychological factors, such as coping styles, may play a role in disability at any level, but are not part of a health condition or health states [24]. In order to gain more insight into their association with HRQoL and/or global QoL, we have clustered the psychological factors into three main groups: mood, beliefs and personality. Mood is a generalized, internal state of feeling (e.g. anxiety, depression and anger) and is closely related to the concepts of affect and emotion. Beliefs refer to people's perceptions of reality including perceptions of health or illness and one's ability to cope with illness (e.g. attitudes, appraisals, religiosity and coping strategies). Personality can be defined as a dynamic and organized set of characteristics which a person possesses and which uniquely influence his or her beliefs, motivations and behaviour in various situations [25].

Data extraction and analysis
We collected information on study characteristics: author, country, sample size and study design, and patient characteristics: age at inclusion, the time of assessment since ALS onset, the functional status of patients (using the Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS), the diagnostic criteria and the type of ALS onset (spinal vs. bulbar). Furthermore, measures of global QoL and HRQoL and of psychological factors, as well as associations between psychological factors and QoL (Health-related and global) were extracted.
Psychological factors were considered "consistent related" if (1) the majority (>50 %) of all studies reported statistically significant bivariate and/or multivariate associations; (2) the majority of the bivariate associations were moderate or strong; and (3) the methodological quality of these studies was adequate or high.

Description of studies included
The search strategy produced a total of 1040 articles (Fig. 1). After removing 153 duplicates, a further 830 articles were removed after screening title and abstract. Agreement on selection of titles and abstracts between the two raters was high (Cohen's kappa 0.82). A total of 57 articles remained for full-text screening; 22 articles met all inclusion criteria. The screening of reference lists produced one additional article [10]. In two studies [16,26], the same cohort data was used; we included the study by Bremer because of a higher quality assessment.
The characteristics of the 22 included studies are presented in Table 1. Studies were published between 1999 and 2015; most were cross-sectional (n = 16); six used longitudinal data [16,[27][28][29][30][31]. The median sample size was n = 49 (range . Studies concerned patients with a mean time of ALS onset between 11.7 months and 5.7 years; the disease severity ranged from 17.4 (severely impaired) to 35.1 points (moderately impaired). The mean age at inclusion varied between 55.3 and 64.0 years; a minority of the patients (7-33 %) had a bulbar onset of ALS, and there was a slight male predominance (M:F ratio1 .5:1). These findings were consistent with those of the general ALS population (mean age 58-63 years; bulbar onset of 30 % and M:F ratio~1.2-1.5:1) [32,33].
Across the studies, fifteen different psychological factors were assessed using 24 different measures. The various instruments for assessing psychological factors are described in Table 2. Mood was investigated in 14 studies, beliefs in 11 and personality in one study. Two ALS-specific questionnaires, the ALS Depression Inventory (ADI-12) [34] and the Motor Neuron Disease Coping Scale (MNDCS) [35] were applied in one [28] and three [30,36,37] studies, respectively. The modified versions of the Hospital Anxiety and Depression Scale (HADS), which were intended not to rely on measuring the physical components of depression, were used in two studies [37,38] (Table 2).
A total of 12 different QoL measures were used in the selected studies, including five different Global QoL measures and seven different Health-related QoL measures (Table 3 and Table 4, resp.). One ALS-specific HRQoL questionnaire, the Sickness Impact Profile ALS (SIP/ALS-19) [39], was used in two studies [14,31].
The average methodological quality score of the studies was 5.3 and ranged from 3 to 8 out of a maximum 8 points (Table 5). Seven studies (32 %) achieved a "high quality" score (≥6/8). Inter-rater agreement on quality of the individual studies was high (ICC = 0.90).

Psychological factors associated with QoL in ALS
An overview of the bivariate and multivariate associations between psychological factors and QoL is presented in Table 6. Due to the heterogeneity of instruments used in assessing both psychological factors (n = 24) and QoL (n = 12), a meta-analysis was not possible.

Mood associated with QoL
Concerning mood, relationships between QoL and anxiety, depression and confusion-bewilderment were found.
Two out of three studies, including one high quality study [37] which used a modified HADS, showed    7 To assess psychological distress in medically ill patients. The instrument concentrates on the psychic rather than the somatic symptoms of mood disorder in order to provide an assessment of mood independent of levels of physical disability in patients with medical illnesses.  To assess a major depressive disorder. Score interpretation: 5 out of 9 symptoms have to be present and represent a change from previous functioning; at least one of the symptoms is either (1) depressed mood or (2)   Generic [46] Personality traits NEO Five Factor Inventory (NEO-FFI) Costa (1992) [78] 60 To assess the five dimensions of personality, postulated by the five-factor model of personality: neuroticism, extraversion, openness, agreeableness, and conscientiousness.
5-point scale. Each of the five-factor subscales consists of 12 items, resulting in mean factor scores ranging from 0 to 4. Score interpretation: Higher score reflects a type of personality.
Generic [27] van Groenestijn et al. Health and Quality of Life Outcomes (2016) 14:107 significant relationships with HRQoL; anxiety was strongly (−0.53) negatively associated with HRQoL. A similar contribution was obtained from multivariate analyses [19]. In contrast, one high quality study failed to find any relationship between anxiety and HRQoL [40]. Significant negative associations with global QoL were found in one out of three studies [41]. More specifically, low trait and state anxiety was associated with higher global QoL and was found in the diagnostic as well as the follow-up phase [41]. Two studies, of which one used a modified HADS, did not find any significant associations with global QoL [15,19].
A significantly negative association with depression and HRQoL was reported in seven out of eight studies, including two of high quality, of which one used a modified HADS. Depression was moderately to strongly (−0.430; −0.60; −0.617) [10,37,42] correlated with HRQoL [10,27,[42][43][44]. The contribution of depressive symptoms to HRQoL was also endorsed in regression analysis [19,27,37,43,44]. A single study of high quality failed, however, to find a significant association with depression and HRQoL in bivariate correlations [40]. Another study used four different HRQoL measures and found two out of four significant associations between depression and HRQoL in bivariate and multivariate analysis [44].
A significantly negative association with depression and Global QoL was reported in four out of eight SEIQoL assesses overall subjective QoL as judged by the patient in healthy or ill individuals. It is derived from a decision analysis technique known as judgement analysis, administered through a semi-structured interview. Patients rate their satisfaction with areas of their life by assessing three aspects of QoL.
The patients have to 1) nominate the life areas (cues) which are important to their QoL; 2) rate their current level of functioning in each of these salient areas; and 3) rate the relative importance of each of their chosen cues. SEIQoL index score: the SEIQoL scores are entirely person-specific, for the purpose of group analyses an overall global or index QoL score (also referred to as a total QoL score) is calculated. The resulting SEIQoL index ranges from 0 (worst possible QoL) to 100 (best possible QoL).
[ 15,42] The Schedule for the evaluation of Quality of Life-Direct Weighting (SEIQoL-DW) Hickey (1996) [81] 15 SEIQoL-DW is a shorter, direct-weight (DW) version of the SEIQoL, employs an alternative method of deriving cue weights using a colored disk. SEIQoL-DW index score: The SEIQoL-DW scores are entirely person-specific; for the purpose of group analyses an overall global or index QoL score (also referred to as a total QoL score) is calculated. The resulting SEIQoL index ranges from 0 (worst possible QoL) to 100 (best possible QoL). [13,28,30,38] The McGill Quality of Life Questionnaire (MQOL) Cohen SR (1995 [82,83] 17 MQOL assesses overall subjective QoL as judged by the patient. Subjects evaluate their lives over the past 2 days on five subscales using a 10-point semantic-differential format. Originally designed for cancer and HIV patients. It is not heavily weighted toward physical function and it includes an existential element. MQOL includes five domains, two of which are health-related: Physical Symptoms (MQOL-Ph) (3 items) and Physical Well-being (MQOL-PW) (1 item); and three are non-health related: Psychological symptoms (MQOL-Ps) (4 item); Existential Well-being (MQOL-EW) (6 items) and Social Support (MQOL-Su) (2 items). Scores on the subscales range from 0 (worst) to 10 (best). The MQOL total score is the mean of the 5 subscales, score ranges from 0 (worst QoL) to 10 (best QoL). MQOL-SIS: besides the subscales there is also a Single-Item Score (SIS): the patient is asked to indicate his/her self-perceived overall QoL in the past two days in a single-item scale (SIS) measuring overall subjective QoL, rated from 0 (very bad) to 10 (excellent). [13,14,16,27,29,41] QoL-single-item question Self-developed by Ganzini (1999) [45] 1 A single-item question to assess patients self-perceived overall QoL. End-points labelled 1 = "my quality of life is as good as it can be" and 6 = "my quality of life is very bad, horrible". [45] QoL-single-item question Self-developed by Krampe (2008) [27] 1 A single-item question to assess patients self-perceived overall QoL. "Over the past seven days, the quality of my life has been": very poor (0)excellent (10).
[27] A prospective long-term follow-up study reported the relationship between depression and HRQoL and global QoL during the first year after baseline measurement. Patients who were more depressed had lower HRQoL and global QoL scores at month 1 and during a 12month follow-up. The results of a linear mixed model analysis showed no interaction effect between depression and time, indicating that more depressed patients did not differ from less depressed patients as far as the trajectories of Global QoL and HRQoL were concerned [27].

Confusion-bewilderment
One study examined the relation between 'confusion-bewilderment' and QoL. In regression analysis, this mood state made a significant positive contribution to HRQoL (β = 0.33) [19].

Beliefs associated with QoL
With regard to beliefs, relationships between QoL and religiosity, spirituality, mindfulness, coping, hopelessness, perception of burden and cognitive appraisal were found. Items are rated on a 5-point scale (low score of 1 to high score of 5) to determine a raw item score. Subsequently, the mean score for each domain is calculated, resulting in a mean score per domain that is between 4 and 20. Finally, this mean domain score is then multiplied by 4 in order to transform the domain score into a scaled score, with a higher score indicating a higher QoL. [19,37] Quality

Religiosity
Five studies assessed the relationship between religiosity and QoL [13,14,16,31,46]: three of these reported HRQoL, four global QoL and two studies both HRQoL and global QoL outcomes. Two out of three studies did not find any significant relationships between religiosity and HRQoL [14,46]. Regression analyses of a third study, which was of high quality, [31] revealed that a high level of religiosity made a significant positive contribution to HRQoL at 6 months' follow-up, but not at the earlier assessment (3 months' follow-up).
Three out of four studies [13,14,16] showed that a higher level of religiosity was significantly related to higher global QoL. Both 'religiosity' and 'private religiosity' (how religious patients perceived themselves to be and the amount of strength and comfort obtained from religious practices) developed a significant, moderate to strong association with global QoL over time (3-16 months' and 6-16 months' follow-up, respectively) [16]. Regression analysis confirmed this increasing relationship between 'private religiosity' and global QoL with time and showed an increase in explained variance of 16 % at 12 months' follow-up [16]. On the other hand, one high quality study did not find any significant associations with religiosity and global QoL at 3 or 6 months' follow-up [31].

Spirituality
Two studies tested the correlation between spirituality and QoL [16,47]. One used an HRQoL measure [47], whereas the other a global QoL measure [16].
The first study [47] split spiritual well-being along the dimensions of religious well-being (which refers to a relationship with God or what is understood as a spiritual being) and existential well-being (which involves a sense of purpose and meaning in life as a means of feeling connected to the world, separate from any specifically religious reference, beliefs and needs). Existential well-being was not associated with HRQoL. In contrast, religious well-being was strongly associated (−0.99) with higher HRQoL, independent of the clinical phase of ALS.     The second study [16] showed that spirituality (which refers to a search for the sacred or divine through any type of life experience) was strongly associated (0.54) with higher global QoL at long-term follow-up (12-16 months) but not at the earlier assessments.

Mindfulness
One recent high quality study found a positive association between mindfulness 'the process of actively making new distinctions about a situation and its environment, or its current context, rather than relying on previous categorizations from the past' [48] and global QoL [29]. The results of a linear mixed model analysis showed that high mindfulness at baseline predicted significantly higher global QoL scores after four months [29].

Coping
Four studies, including three high quality studies, investigated the associations between coping and QoL [30,36,37,46], in which two studies reported HRQoL and two studies global QoL measures.
One high quality study out of two showed that 'adoption of positive coping strategies' was moderately positively (0.46) associated with HRQoL [37]. The second study [46] related 14 coping strategies to HRQoL (36-item Short Form (SF36); Mental Component Summary score (MCS) and Physical Component Summary scores (PCS)). Of these 28 bivariate correlations, three were significantly associated: negative, moderate correlations were noted between MCS and substance use (−0.44) and between MCS and venting (an externalizing coping technique, the outward expression of emotions) (−0.38). PCS was positively, moderately associated with emotional support (0.38). The other 11 coping strategies (e.g. acceptance, denial, selfblame) were not associated with HRQoL [46].
Two high quality studies, analyzing the same cohort, showed that the coping strategy 'problem management' was positively associated with global QoL. The first study [36] with a cross-sectional design, using multivariate regression analysis, revealed positive associations between the coping strategy 'problem management' and 'emotional avoidance' and global QoL at baseline [30]. Analysis of a longitudinal follow-up [30] revealed that only the coping strategy 'problem management' was a significant predictor; patients who searched more frequently for information and support at baseline reported higher global QoL at 3 to 6 months' follow-up [30].

Hopelessness
One study tested the association of hopelessness and global QoL [45]. It was shown that greater hopelessness was moderately correlated with lower global QoL (0.43). This relationship was still significant in a multivariate regression analysis with control variables.

Perception of burden to others
A single study examined the 'perception of burden to others'. Having the belief of being a burden to others was moderately associated with lower global QoL (0.45). The association remained significant in the regression analyses [45].

Cognitive appraisal
A single study [36] assessed 'cognitive appraisal' , which was split into patient's primary (motivational relevance, motivational congruence) and secondary appraisal (problem focused and emotional focused coping potential) and related to global QoL. Results of the regression analysis showed no associations of cognitive appraisal with global QoL. The variance of the global QoL scores could not be significantly accounted for by any of the appraisals scales.
Personality associated with QoL Neuroticism, extraversion, openness, agreeableness and conscientiousness One study [27] investigated the relationship between personality factors and QoL. In the regression analysis, it was shown that among the five personality factors (neuroticism, extraversion, openness, agreeableness and conscientiousness), only agreeableness had a strong positive association with both global QoL and HRQoL. Agreeableness refers to 'a personality trait manifesting itself in individual behavioral characteristics that are perceived as kind, sympathetic, cooperative, warm and considerate' [49]. There was also a significant interaction effect of agreeableness and time, meaning that agreeableness significantly influenced the course of global QoL and HRQoL; patients who scored higher on agreeableness had higher QoL ratings at baseline measurement but their decline in QoL was steeper compared to patients with lower scores in agreeableness [27].

Discussion
The aim of the present review was to systematically collect and appraise evidence of the relationships between psychological factors (mood, beliefs, personality) and QoL in patients with ALS. This review showed that higher levels of anxiety and depression appeared to be related to a poorer HRQoL, whereas a higher level of religiosity seemed to be related to higher global QoL. Furthermore, associations might change during the disease course.

Mood
Anxiety seemed to be negatively related to HRQoL, because higher levels of anxiety were consistently associated with a poorer HRQoL. In contrast, global QoL showed no associations with anxiety; the association could not, however, be refuted because of one poor quality study. Depression was negatively associated with HRQoL, suggesting that the presence of depressive symptoms is related to a poorer HRQoL. On the other hand, for global QoL, we could not support a relationship with depression, because of inconsistent results.
Mood appeared to be related to HRQoL but not to global QoL. This is in concurrence with cancer studies [9,50], which also revealed that depression explained a large amount of variation in HRQoL, but not global QoL. Our results might in part be ascribed to conceptual overlap [51] between determinants and outcomes. For example, questions about feelings of anxiety and depression are often also included in a HRQoL measure (e.g. "Have you felt downhearted and blue?" SF-36; question 9f [52]), and so studying anxiety and depression as determinants of HRQoL may result in strong associations between determinants and outcomes. This contamination is less likely between mood and global QoL measures, because global QoL assesses such a wide spectrum of domains that contribute to overall QoL as judged by the patient.
Furthermore, it is important to be aware that there is recent evidence suggesting that depression questionnaires, specifically the Beck Depression Inventory (BDI) [53] and to a lesser degree the Hospital Anxiety and Depression Scale (HADS) [54], tend to overestimate depression in ALS, since these scores are highly influenced by the physical impairment of the patients [55,56]. Consequently, the relationships that have been found between depression and HRQoL is questionable.

Beliefs
There seemed to be no relationship between religiosity and HRQoL, because most studies showed weak and nonsignificant associations. However, religiosity appeared to be positively associated with global QoL in the majority of the studies, including one high quality study. Consequently, we support the assumption that a high level of religiosity made a significant positive contribution to better global QoL. However, these results might also point out to contamination in concepts between religiosity and global QoL since the The McGill Quality of Life Questionnaire (MQOL) total score includes items about existential well-being.
Religiosity might be important for the individual's global QoL because it may create meaning and coherence when an individual's world is devastated by a distressing and progressive disease [47]. These findings are mirrored in other diseases such as Multiple Sclerosis [57] and advanced cancer [58]. It should be taken into account that most of the included studies about religiosity are from North America and the religiosity questionnaire which was used relies predominantly on monotheistic terminology, about belief in God or experience of God [59]. In current (western) culture, people are more interested in spirituality [60] and mindfulness [55] and are searching for a connection with the divine within themselves, instead of a connection with an external almighty power [60]. The fact that religiosity and spirituality are culture dependent and are defined differently in each country might explain the heterogeneity in findings on association of QoL.

Personality
The search has yielded only one hit concerning personality factors, suggesting that personality factors were not considered to be the most important psychological factors influencing QoL. Only a single study of low quality was included, conclusive associations between HRQoL or global QoL could, therefore, not be established.

Miscellaneous
Several other psychological factors were reported in only a single study or measured twice in the same cohort [30,36]: there was, therefore, insufficient evidence to support associations between HRQoL or global QoL and the following psychological factors: confusion-bewilderment (mood); spirituality, mindfulness, coping styles, hopelessness, perception of burden, cognitive appraisal (beliefs) and neuroticism, extraversion, openness, agreeableness and conscientiousness (personality).

Associations throughout the disease course
Although there was not enough evidence per psychological factor, it is valuable to point out with regard to psychological factors as a whole, that associations might change throughout the course of the disease; religiosity and spirituality appeared to become more positively associated with global QoL over time. This is in accordance with the theory of Waldron [61] who suggested that psychological adaptation to terminal illness may involve a shift in focus of determinants of QoL; in the initial stages of a progressive illness, patients may focus on physical functioning and on decreasing disability, but as the illness progresses, the importance of these issues may be replaced by a focus on the psychosocial and spiritual domains.

Strengths and limitations of this systematic review
This is the first systematic review on associations between psychological factors and QoL in patients with ALS. This review was carried out in accordance with the PRISMA guidelines (see online Additional file 1). The methodological quality of the studies and the consistency of the associations between psychological factors and HRQoL and global QoL were comprehensively appraised. A limitation, however, of our study is that only a small number of psychological factors could be compared, because most of the associations with QoL were only reported once. Besides, our review may have missed relevant papers published in non-English journals. Finally, as studies with significant results are more likely to be published than studies without significant results, publication bias has to be taken into account.

Limitations of the literature
First, the heterogeneity of the literature with respect to instruments for assessing psychological factors and QoL may have influenced the associations. Levels of anxiety and depression, for example, were measured using 5 and 7 different questionnaires, respectively, and moreover, with a mix of both generic and ALS-specific questionnaires, or questionnaires modified for ALS. Concerning QoL measures, 7 different HRQoL and 5 different global QoL (mostly generic) measures, were extracted. Second, in order to detect which psychological factors affect HRQoL and global QoL over the course of the disease, it is essential to cluster the data of patients according to the same disease stage (diagnostic stage, rehabilitation stage and terminal stage [62]). In fact, only one study analyzed the determinants of patients in the diagnostic phase separately [41]. Other longitudinal studies only reported changing associations of depression, coping, religiosity and personality factors after baseline measurement, without specific information about the disease stage. Other limitations concern studies with a crosssectional design and the overrepresentation of small studies, without an adequate sample size in relation to the number of determinants.

Conclusions, clinical implications and further research
Our results suggest that higher levels of anxiety and depression are related to a poorer HRQoL, whereas higher levels of religiosity appeared to be related to better global QoL. Furthermore, associations might change throughout the disease course.
Therefore it is important for health professionals to become aware of the relationships between psychological factors and QoL, as these relationships identify possible targets for interventions to improve QoL. It seems relevant for health professionals in ALS care, to focus on influencing mood and beliefs in order to improve HRQOL and global QoL. Furthermore, it is relevant to make a distinction between HRQoL and global QoL, because HRQoL is expected to decline, according to a decrease in mental and physical functioning, whereas global QoL seems more dependent on other factors, such as existential concerns.
More high quality research is needed to confirm the assumed association between anxiety, depression and religiosity and HRQoL and global QoL and to investigate how and when these factors can be targeted in ALS care. Coping, spirituality, mindfulness, hopelessness, perception of burden and agreeableness might be other promising factors that influence QoL, but warranted further investigation. More longitudinal studies in larger samples are needed because they allow causal relationships and effects of time to be identified. Furthermore, uniformity of measures for QoL and psychological factors, preferably ALS-specific, are required in order to obtain reliable, comparable data. As small sample sizes are inherent to ALS research, the answer may lie in international collaboration and data gathered by online survey.