Our study has identified that behavioral changes (e.g. disinhibition, impulsivity) and caregivers’ level of stress have greater impact on caregiver burden than any other patient variables, including level and pattern of physical disability.
Behavioral symptoms, as measured by the Frontal Systems Behavioral Scale (FrSBe), have previously been reported to be associated to higher caregiver burden,
[16, 18] although only one study has considered variables related to caregivers
. We have confirmed this association using the Cambridge Behavioural Inventory - Revised and, importantly, found that the strongest predictor of caregiver burden was the patient’s abnormal behavior. It should be noted that, as in other studies, apathy was the most common behavioral change
[2, 18, 29], but this factor did not enter the model of caregiver burden. Stereotypical behaviors were also reported at similar levels to abnormal behavior, but again, this factor did not explain caregiver burden. Specific changes in behavior, notably disinhibition and impulsivity, are particularly distressing for caregivers. We suspect that many clinicians involved in the management of ALS patients are unaware of the high rate of these changes and their impact on family relationships. Specific interventions targeting carers and developing their coping skills have been shown to be successful in FTD
 and may have a successful place in the management of ALS dyads.
A number of studies have shown that caregivers of patients with ALS report high levels of burden, anxiety and depression in association with the patient’s loss of physical functions
[14, 15]. In our study, somewhat surprisingly, burden was not associated with changes in physical function, as measured by the ALSFRS-R. One possible explanation for this lack of association may be that behavioral changes which overshadow the lesser influence of physical disability in a log regression model. Moderate or worse depressive symptoms were present in 23% of caregivers, with similar rates of stress. These symptoms did not, however, enter the model explaining caregiver burden. The reported rate of caregiver depression in caregivers of people with ALS has varied from 11 to 61%
[31, 32] which almost certainly reflects the use of different instruments, the threshold used to determine depression and the level of disability of patients in the cohorts. It is interesting that caregiver depression had lesser role in the genesis of caregiver burden than seen in pure FTD, despite the overlap in behavioral symptoms between bvFTD and ALS
The levels of anxiety and stress (around 20%) in our caregivers were lower than reported previously
[33, 34]. Again, the variability probably reflects the use of different instruments. Caregivers’ state anxiety has been previously related to their trait anxiety more than external factors
. Stress rather than depression and anxiety was the main psychological caregiver variable contributing to caregiver burden. This is clearly relevant to management since can be addressed in MND clinics and are potentially amenable to intervention.
A longitudinal study identified that, initially, caregiver distress was best predicted by the psychosocial impact of ALS symptoms and emotional lability. Later, caregiver distress related to satisfaction with social relationships and social support
 rather than behavioral changes, although these were limited to dysexecutive symptoms rather than the FTD-like features explored by the CBI. Future longitudinal studies should investigate the role of specific behavioral changes, caregiver variables, and disease progression especially given the fact that caregiver burden seems to increase over time. A limitation of this study was the nature of the survey via the MND Association, where carers who responded to the invitation to the research study were likely to be those willing to respond the questionnaires. Still, our results corroborated those published previously, and the patients’ characteristics mirrored those in large cohorts.
The fact that ALS patients presented similar behavioural symptoms to patients with FTD has been reinforced by a recent genetic discovery. The presence of a hexanucleotide repeats in the intron of C90RF72 in almost a quarter of cases with familial ALS and 12% of the cases with familial FTD
[35, 36] support strongly the idea of the continuum between ALS and FTD.