Article Text

Download PDFPDF

Original research
Relationship between neuropsychiatric disorders and cognitive and behavioural change in MND
  1. Caroline A McHutchison1,2,3,
  2. Danielle Jane Leighton2,4,
  3. Andrew McIntosh3,5,
  4. Elaine Cleary6,
  5. Jon Warner6,
  6. Mary Porteous6,
  7. Siddharthan Chandran2,4,
  8. Suvankar Pal2,4,
  9. Sharon Abrahams1,2
  1. 1 Human Cognitive Neurosciences, Department of Psychology, The University of Edinburgh, Edinburgh, UK
  2. 2 Euan MacDonald Centre for MND Research, The University of Edinburgh, Edinburgh, UK
  3. 3 Centre for Cognitive Ageing and Cognitive Epidemiology, The University of Edinburgh, Edinburgh, UK
  4. 4 Centre for Clinical Brain Sciences, The University of Edinburgh, Edinburgh, UK
  5. 5 Division of Psychiatry, Royal Edinburgh Hospital, Edinburgh, UK
  6. 6 Centre for Genomic & Experimental Medicine, South East Scotland Genetics Service, Western General Hospital, Edinburgh, UK
  1. Correspondence to Ms Caroline A McHutchison, Human Cognitive Neurosciences, Department of Psychology, The University of Edinburgh, Edinburgh, Edinburgh, UK; Caroline.McHutchison{at}ed.ac.uk

Abstract

Objective In this population-based study, we aimed to determine whether neuropsychiatric history, medication or family history of neuropsychiatric disorders predicted cognitive and/or behavioural impairment in motor neuron disease (MND).

Methods People with MND (pwMND) on the Scottish Clinical, Audit, Research and Evaluation of MND (CARE-MND) register, diagnosed from January 2015 to January 2018, with cognitive and/or behavioural data measured using the Edinburgh Cognitive and Behavioural ALS Screen were included. Data were extracted on patient neuropsychiatric, medication and family history of neuropsychiatric disorders. We identified patients with cognitive impairment (motor neuron disease with cognitive impairment (MNDci)), behavioural impairment (motor neuron disease with behavioural impairment (MNDbi), both (motor neuron disease with cognitive and behavioural impairment (MNDcbi)) or motor neuron disease–frontotemporal dementia (MND-FTD).

Results Data were available for 305 pwMND (mean age at diagnosis=62.26 years, SD=11.40), of which 60 (19.7%) had a neuropsychiatric disorder. A family history of neuropsychiatric disorders was present in 36/231 (15.58%) of patients. Patient premorbid mood disorders were associated with increased apathy (OR=2.78, 95% CI 1.083 to 7.169). A family history of any neuropsychiatric disorder was associated with poorer visuospatial scores, MNDbi (OR=3.14, 95% CI 1.09 to 8.99) and MND-FTD (OR=5.08, 95% CI 1.26 to 20.40). A family history of mood disorders was associated with poorer overall cognition (exp(b)=0.725, p=0.026), language, verbal fluency and visuospatial scores, and MND-FTD (OR=7.57, 95% CI 1.55 to 46.87). A family history of neurotic disorders was associated with poorer language (exp(b)=0.362, p<0.001), visuospatial scores (exp(b)=0.625, p<0.009) and MND-FTD (OR=13.75, 95% CI 1.71 to 110.86).

Conclusion Neuropsychiatric disorders in patients and their families are associated with cognitive and behavioural changes post-MND diagnosis, with many occurring independently of MND-FTD and C9orf72 status. These findings support an overlap between MND, frontotemporal dementia and neuropsychiatric disorders, particularly mood disorders.

  • neuropsychiatric disorders
  • cognition
  • behaviour
  • motor neuron disease
  • amyotrophic lateral sclerosis
  • frontotemporal dementia
View Full Text

Statistics from Altmetric.com

Introduction

Up to 50% of people with amyotrophic lateral sclerosis (ALS), the most common form of motor neuron disease (MND), experience cognitive and/or behavioural changes, with 15% meeting the criteria for frontotemporal dementia (FTD).1 2 These changes represent a range of frontotemporal spectrum disorders in ALS (ALS-FTSD).3 A genetic and pathological overlap between MND and FTD has been established with the C9orf72 repeat expansion occurring in ~6.3% of sporadic ALS cases, ~5.8% of sporadic FTD cases4 and ~33% of familial MND cases,5 and the TAR-DNA-binding protein 43 occurring in almost all ALS, ALS-FTD and ~45% FTD cases.6–8

Cognitive and behavioural symptoms in MND range from no impairment to cognitive impairment, behavioural impairment, both or ALS-FTD.3 An association between neuropsychiatric disorders and ALS has previously been shown.9 10 Neuropsychiatric symptoms including schizophrenia, depression, anxiety, and drug abuse or dependence are common in the years prior to and following diagnosis in both ALS10 11 and behavioural-variant FTD.12 Genome-wide association studies have revealed a modest genetic correlation between schizophrenia and ALS.13 Furthermore, higher rates of psychosis, suicidal behaviour14 and autism15 have been shown in the first-degree and second-degree relatives of patients with ALS,14 15 particularly in their children.11 Despite evidence of an overlap of ALS-FTSD and neuropsychiatric disorders, it is unclear if neuropsychiatric disorders also contribute to the spectrum of cognitive and behavioural changes in ALS.

Using population-based data from the Scottish Clinical, Audit, Research and Evaluation of MND (CARE-MND) register, this study aimed to look at the association between cognitive and behavioural changes in MND as measured using the Edinburgh Cognitive and Behavioural ALS Screen (ECAS)16 and neuropsychiatric disorders in MND kindreds. Specifically, we aimed to determine if the presence of cognitive and behavioural changes can be predicted by the patient’s

  1. History of neuropsychiatric disorders.

  2. History of neuropsychiatric medication.

  3. Family history of neuropsychiatric disorders.

Methods

Participants

We studied people with MND (pwMND) registered on the CARE-MND platform, a population-based, prospective database of patients in Scotland diagnosed with MND.17 18 Since 2015, a major update of the register has taken place. Anonymised data for collaboration can be obtained on reasonable request from the CARE-MND register. In Scotland, any patient receiving a diagnosis of MND (including ALS and subtypes such as primary lateral sclerosis (PLS), progressive muscular atrophy (PMA) and progressive bulbar palsy) is included on the register. pwMND can provide consent for their anonymised information to be used for research purposes. This study includes all consenting patients diagnosed with MND (240 ALS, 23 PLS, 9 PMA and 23 ALS-FTD) between January 2015 and January 2018 with ECAS data available.

Measures

History of neuropsychiatric disorders

Neuropsychiatric history in patients

We identified patients with a history of neuropsychiatric disorder (pre-MND or post-MND diagnosis) from data on the CARE-MND register, using medical history information. We assessed each patient on a case-by-case basis and identified any relevant neuropsychiatric diagnoses (table 1). Each diagnosis was then categorised based on the International Classification of Diseases, 10th Revision (ICD-10).19 We created a summary variable to indicate (yes/no) the presence of a relevant neuropsychiatric diagnosis for each individual.

Table 1

Neuropsychiatric disorders of patients with MND on the CARE-MND register

Neuropsychiatric medication history in patients

We assessed each patient’s medication history (pre-MND and post-MND diagnoses) to identify any relevant neuropsychiatric medication (online supplementary eTable 1). We excluded medication typically used to treat the symptoms of MND (eg, lorazepam) and cases where alternative or secondary use of the medication were stated (eg, muscle cramps). We created a summary variable to indicate (yes/no) the presence of a relevant neuropsychiatric medication for each individual.

Supplementary data

Family history of neuropsychiatric disorders

Information on a family history of neuropsychiatric disorders (pre-MND diagnosis) was available on the CARE-MND register. These data were collected during routine clinical assessment around the time of MND diagnosis using a standardised proforma. Patients provided information on any family history of MND, dementia, and neurological or neuropsychiatric disorders across four generations of blood relatives (patient’s grandparents to children). We defined a positive family history as one or more relatives with a diagnosed neuropsychiatric disorder.

Cognition and behaviour

ECAS cognitive and behavioural scores

Cognition and behaviour were measured using the ECAS.16 This brief assessment, specifically designed for patients with motor degeneration, is both sensitive and specific to the types of impairments shown by patients with ALS20–23 and is routinely undertaken in Scotland. ECAS subscales are grouped into ALS-specific domains (language, verbal fluency and executive functioning) and ALS-non-specific domains (memory and visuospatial skills). A total score (max 136) and domain-specific scores (ALS-specific, max 100; ALS-non-specific, max 36) are calculated with established cut-offs indicating global and domain-specific impairments. The ECAS has been shown to be a sensitive and valid measure of cognition and behaviour in MND16 20 with good convergent validity with other screening tests.21–24

The ECAS behavioural interview was used to assess behavioural change across five domains based on behavioural-variant FTD diagnostic criteria25: disinhibition, apathy or inertia, loss of sympathy/empathy, perseveration and eating behaviour/hyperorality. The interview was completed by an informant, with a score of 1 indicating a behavioural change for that domain, with a maximum of 5 denoting the number of behavioural domains affected.

MND-FTSD subclasses

Using all available cognitive and behavioural data, we classified patients using the ALS-FTSD criteria.3 We used published cut-offs for the ECAS total and ALS-specific scores,16 which have been shown to have maximum sensitivity and specificity to detect cognitive impairment against extensive gold standard neuropsychological assessment.20 As we included all MND subtypes, we referred to these subclasses as MND-FTSD. Using this method, we identified those with cognitive (motor neuron disease with cognitive impairment (MNDci)), behavioural (motor neuron disease with behavioural impairment (MNDbi)) or both cognitive and behavioural impairment (motor neuron disease with cognitive and behavioural impairment (MNDcbi)). Cases of MND-FTD were identified on the CARE-MND register based on a diagnosis given by a neurologist during clinical examination.

The number of patients in some MND-FTSD subclasses was small for some analyses. To increase power, we collapsed the categories to identify patients with any type of cognitive (MNDci, MNDcbi and MND-FTD) or behavioural (MNDbi, MNDcbi, MND-FTD) impairment.

Patients recruited to the CARE-MND register donated blood or saliva for genetic analysis. C9orf72 genotyping was carried out using PCR based on methods previously described.26

Statistical analysis

Some participants had multiple ECAS scores available from repeat testing (n=39, max=3 entries). We included entries with the most complete data (n=23) and where the date of assessment was available (n=2). In cases where there were two entries within 6 months of each other (n=5), we used the first assessment to avoid potential practice effects.27 28 We examined predictors of cognitive and behavioural change in MND, and given that these may increase during later stages of the disease,29 we used the most recent ECAS entry for the remaining cases (n=9).

Dates of neuropsychiatric diagnosis and prescription of medication were available for some but not all patients on the CARE-MND register. We focused our analyses on all identified cases to increase statistical power. In addition, we conducted secondary analyses examining only those with confirmed premorbid information based on the date of diagnosis or prescription of medication.

We examined the relationships between neuropsychiatric disorders, medication and family history and two outcomes: (1) continuous cognitive and categorical (yes/no) behavioural ECAS scores and (2) MND-FTSD subclasses, including the presence of any cognitive or behavioural impairment. The use of continuous scores increased statistical power to detect effects. However, the use of MND-FTSD subclasses to identify predictors of cognitive and/or behavioural impairments has important clinical implications. For analyses using continuous outcomes, we used linear regression and reverse log-transformed ECAS scores where appropriate to avoid violation of normality. Results were back transformed for reporting. We examined the presence of each behavioural change using a series of binomial logistic regressions and MND-FTSD subclass using multinomial regression. For collapsed MND-FTSD subclasses indicating the presence of any cognitive or behaviour impairment, we used binomial logistic regression. Where appropriate, sensitivity analyses were conducted to determine if the associations were driven by MND-FTD cases and whether they remained after controlling for C9orf72 status. All analysis was conducted in R,30 and power calculations were conducted using the pwr package.31

Results

Information was available for 305 pwMND with ECAS data (mean age at diagnosis=62.26 years, SD=11.40; table 2). A total of 211 (69.2%) had complete cognitive and behavioural data, 82 (26.9%) had some cognition only and 12 (3.9%) had behavioural data only. C9orf72 status was available in 259 (84.9%) patients, of which 24 (9.3%) carried disease-causing expansions.

Table 2

Descriptive statistics of patients with MND on the CARE-MND register with ECAS data (N=305)

Total ECAS scores (n=292, unavailable for one case) ranged from 24 to 132 out of 136 (mean=104.74, SD=18.48), and apathy was the most common behavioural change (68/229, 29.7%; see figure 1 for frequencies of impairment). A MND-FTSD syndrome was found in 156 (51.1%): MNDci was present in 77 (25.2%), MNDbi in 33 (10.8%) and MNDcbi in 29 (9.5%) patients. MND-FTD was identified in 17 (5.6%) patients based on a neurologist’s diagnosis. The remaining 149 (48.9%) patients did not fulfil the criterion for cognitive and/or behavioural impairment; however, 18 patients were impaired on the ALS-non-specific but not the ALS-specific subscales. There were no significant differences between those with and without any impairment on demographic or clinical characteristics.

Figure 1

Frequency of impairment on the ECAS. ALS, amyotrophic lateral sclerosis; ECAS, Edinburgh Cognitive and Behavioural ALS Screen.

The relationship between patients’ neuropsychiatric history and cognition and behaviour

Of the 305 people with ECAS data on the CARE-MND register, 60 patients had a medical history of a neuropsychiatric disorder. Those with and without a neuropsychiatric disorder did not differ on C9orf72 status (χ2=0.098, p=0.755); however, those with a neuropsychiatric disorder had a shorter disease duration at the time of ECAS (median=81 days) than those without (median=177 days; W=8458, p=0.005). No relationship was found between disease duration and total ECAS, ALS-specific or ALS-non-specific scores.

Mood disorders were the most common (n=42) followed by neurotic disorders (n=19), personality disorders (n=3), and schizophrenia, schizotypal and delusional disorders (n=1, table 3). Based on diagnosis date, 39 had a definite premorbid neuropsychiatric disorder (2 were diagnosed post-MND, 19 had missing dates). Those with a premorbid neuropsychiatric disorder had a shorter disease duration at the time of ECAS than those without (77 vs 177 days; W=5583.5, p=0.006). There were no other significant differences on demographic or disease characteristics.

Table 3

Frequency of neuropsychiatric disorders in patients with MND and their family members on the CARE-MND register

The analyses of ECAS scores revealed that a diagnosis of a mood disorder was associated with poorer language scores but no other specific subscale (table 4). A diagnosis of any neuropsychiatric disorder was associated with increased likelihood of apathy but no other specific behavioural domain. Examination of each individual neuropsychiatric disorder showed an association between mood disorders and increased risk of apathy. We did not find any other significant associations between patient neuropsychiatric disorders and cognition or behaviour. After excluding MND-FTD cases, these associations were no longer significant. After controlling for C9orf72 status, the association between mood disorders and language scores was no longer significant; however, all other associations remained.

Table 4

Patient depression and anxiety associations with ECAS cognitive and behavioural scores

In the secondary analysis examining those with confirmed premorbid neuropsychiatric disorders only (n=39), the association between mood disorders and apathy remained (online supplementary eTable 2). The associations between mood disorders and language scores and any neuropsychiatric disorder and apathy were no longer significant; however, in both cases, statistical power was low (0.34 and 0.16, respectively).

The analyses of MND-FTSD classifications revealed that mood disorders were associated with increased odds of MNDbi versus MND with no impairment (table 5). This association remained after excluding MND-FTD cases and controlling for C9orf72 status. We did not find any other associations (table 5 and figure 2A) including those examining premorbid cases only (online supplementary eTable 3 and eFigure 1A).

Table 5

Patient neuropsychiatric disorders and MND-FTSD subclasses

Figure 2

Risk of any cognitive (MNDci, MNDcbi or MND-FD) or behavioural (MNDbi, MNDcbi or MND-FTD) impairment for all identified cases. FH. family history; MNDbi, motor neuron disease with behavioural impairment; MNDci, motor neuron disease with cognitive impairment; MNDcbi, motor neuron disease with cognitive and behavioural impairment; MND-FTD, motor neuron disease–frontotemporal dementia.

The relationship between patients’ history of neuropsychiatric medication and their cognition and behaviour

Of the 305 patients with ECAS data on the CARE-MND register, 127 had a relevant neuropsychiatric medication in their medical history (92 antidepressants, 54 benzodiazepines and 5 antipsychotics). Based on date of medication prescription and date of MND diagnosis, 25 people were identified with premorbid neuropsychiatric medication (57 were prescribed post-MND diagnosis and 45 missing dates). There were no significant differences between those with premorbid neuropsychiatric medication and those without demographic or disease characteristics.

We found no associations between the prescription of any neuropsychiatric medication and cognitive and behavioural changes (online supplementary eTable 4), MND-FTSD subclass (online supplementary eTable 5), or behaviour or cognitive impairment (online supplementary eFigure 2). All results remained non-significant when including only those with medication prescribed prior to MND diagnosis (online supplementary eTables 6–7 and eFigure 1B).

The relationship between patients’ family history of neuropsychiatric disorders and their cognition and behaviour

Of the 305 people with ECAS data on the CARE-MND register, information on family history of neuropsychiatric disorders was available for 231 (75.74%) patients. Patients with and without a family history of neuropsychiatric disorders did not differ on C9orf72 status (χ2=3.209, p=0.073). A positive family history of any neuropsychiatric disorder was present in 36/231 (15.58%). Mood disorder was the most common (n=16) followed by schizophrenia, schizotypal or delusional disorders (n=10), and neurotic disorders (n=9, table 3). A positive family history of dementia was reported in 76/266 (29%, data unavailable for 39) and was associated with increased odds of a family history of neuropsychiatric disorders (OR=4.198, 95% CI =1.996 to 8.958). There were no other differences between those with and without a family history of neuropsychiatric disorders on demographic or disease characteristics.

A family history of any neuropsychiatric disorders was associated with poorer visuospatial scores. A family history of mood disorder was significantly associated with poorer total ECAS and ALS-non-specific scores, as well as poorer language, verbal fluency and visuospatial scores. For a family history of neurotic disorders, we found an association with poorer language and visuospatial scores (table 6). We did not find any other significant associations between family neuropsychiatric disorders and cognition or behaviour. After excluding MND-FTD cases, the association between family history of neurotic disorders and language scores remained (exp(B)=0.463, p=0.013). All other associations were no longer significant. All associations remained after controlling for C9orf72 status.

Table 6

Family history of neuropsychiatric disorders and cognitive and behavioural outcomes in patients

For the MND-FTSD subclass, a family history of any neuropsychiatric disorder was associated with an increased risk of MND-FTD and MNDbi versus MND with no impairment. Examination of each neuropsychiatric disorder showed that a family history of mood and neurotic disorders was associated with an increased risk of MND-FTD versus MND with no impairment (table 7).

Table 7

Family history of neuropsychiatric disorders and MND-FTD subclass

The presence of any behaviour impairment was associated with a family history of any neuropsychiatric disorders or neurotic disorders (figure 2b). We did not find any associations between family history of neuropsychiatric disorders and the presence of any cognitive impairment.

Discussion

We showed that the associations between MND, FTD and neuropsychiatric disorders extend into the cognitive and behavioural features in MND. A premorbid mood disorder in a patient with MND is associated with a 2.78 increase in the odds of apathy post-MND diagnosis, and a family history of any neuropsychiatric disorder is associated with 3.14 and 5.08 increased odds of MNDbi and MND-FTD, respectively. More specifically, a family history of mood disorder is associated with poorer overall cognition, in addition to a 7.57 increase in the odds of MND-FTD. Patients with a family history of neurotic disorders are more likely to have poorer cognition, including language and visuospatial functioning, and a 13.75 increase in the odds of MND-FTD. These associations may contribute to the spectrum of cognitive and behavioural changes in MND.

Previous research has suggested that the C9orf72 expansion accounts for some of the shared predisposition between MND, FTD and neuropsychiatric disorders. In one study, C9orf72 was associated with higher rates of schizophrenia, suicide and autism in ALS and FTD family members.32 Some studies have also shown an association between C9orf72 and cognition in patients with ALS and FTD; however, these findings have been mixed.33 34 We showed that those with and without a personal or family history of neuropsychiatric disorders did not differ on C9orf72 status. To our knowledge, we are also the first to show that neuropsychiatric history of both patient and family is associated with cognitive and behavioural changes in MND, independent of the patient’s C9orf72 status. These findings suggest that, although a shared predisposition exists, factors other than C9orf72 status are contributing to this susceptibility. Furthermore, those with a premorbid neuropsychiatric disorder had a shorter disease duration at the time of ECAS assessment. This suggests that a neuropsychiatric history may be associated with poor cognition in the earlier stages of the disease; however, further research is needed to confirm this.

We showed associations across both ALS-specific and non-specific cognitive domains. Although ALS-non-specific impairment, including visuospatial impairment, are less common in MND, different cognitive profiles exist and other factors may be contributing. Impaired visuospatial skills are common in Alzheimer’s disease (AD)35; therefore, our results may reflect the presence of mild AD pathology. Alternatively, visuospatial skills have been shown to decline with increasing disease stage, even in FTD35; therefore, it may be an effect of disease stage.

Many associations did appear to be driven by patients with MND-FTD and were no longer significant after removal of these cases. However, a family history of neurotic disorders and language scores remained significant after the removal of MND-FTD cases showing that these associations exist in non-demented patients with MND. Furthermore, many associations were found with MNDbi and more specific cognitive deficits, suggesting that neuropsychiatric history is associated with the full spectrum of cognitive and behavioural changes in MND. Future research using more in-depth measures of neuropsychiatric disorders, including subclinical symptoms, may be more sensitive and may reveal further associations.

We were able to identify cases of premorbid neuropsychiatric disorders in patients with MND using diagnosis date and found a strong relationship between premorbid mood disorders and apathy post-MND diagnosis. These findings suggest that a history of mood disorders may be a risk factor for the development of apathy in MND. Apathy or demotivation is the most common behavioural impairment in ALS.36 37 Although some symptoms of depression and apathy may overlap, a dissociation in ALS has been demonstrated,37 and the distinct profile of initiation apathy found in ALS has been related to a cognitive failure to initiate as demonstrated by verbal fluency deficits.38 Early signs of apathy may have been captured as premorbid mood disorders on the CARE-MND platform. Little is known about timing of cognitive and behavioural symptoms relative to motor symptom onset in MND. However, these are likely to be mood disorders as only 7/28 (25%) cases were diagnosed with mood disorders in the 5 years prior to MND diagnosis.

Associations between neuropsychiatric history and later-life disease have previously been demonstrated. An increased risk of dementia has been associated with a history of depression, anxiety, bipolar disorder, schizophrenia and alcohol dependence.39 Parkinson’s disease has also been associated with premorbid depression.40 However, few studies have explored how premorbid neuropsychiatric disorders relate to cognitive and behavioural changes in neurodegenerative disease, including MND. The findings from this study highlight the contribution of neuropsychiatric symptoms to the cognitive and behavioural profiles in MND.

Our population-based study includes all patients with MND in Scotland with cognitive and/or behavioural information assessed using the ECAS. This information allowed us to group patients using established cut-offs and the latest ALS-FTSD classifications.3 20 Rates of cognitive and/or behavioural impairment in our sample (156/305, 51.1%) were consistent with previous research.1 We identified cases of MND-FTD based on neurologist diagnosis; therefore, they may not be in strict adherence to the ALS-FTSD criteria.3 This may have contributed to fewer cases being identified than in previous research; however, our rates (17/305, 5.57%) were comparable to published frequencies.41 A classification of MND-FTD cannot be made using ECAS data available on the CARE-MND platform as information indicating progression of cognitive and/or behavioural symptoms, a key criterion, is unavailable. Based on the other two MND-FTD criteria,3 an additional 23 patients may have MND-FTD; however, this could not be confirmed. The ECAS is generally administered shortly following diagnosis; however, changes in cognition and behaviour have been shown to increase in the final stages of the disease29; therefore, some patients may go on to develop MND-FTD.

Neuropsychiatric disorders were uncommon in the MND kindreds included in our study. We identified mood disorders in 42 (13.77%) patients and 28 (9.18%) family members, which is lower than expected based on previous research14 15 and rates found in the UK Biobank (27.20% of 123 000).42 This resulted in small numbers in some analyses; therefore, these results should be interpreted cautiously. We were unable to examine associations with psychosis, schizophrenia or autism, and neuropsychiatric disorders, which have been shown to occur at higher rates in MND kindreds.11 14 15 The number of patients with neuropsychiatric medication was higher than those with a neuropsychiatric history (127 vs 60). Information on the duration and reason for neuropsychiatric medication use was not available on the CARE-MND register. Therefore, cases were neuropsychiatric medication was prescribed to treat short-term symptoms not meeting diagnostic criteria or non-neuropsychiatric symptoms (eg, duloxetine to treat fibromyalgia) may be been included. This may have contributed to the lack of associations found.

We used patient-reported family history data, which may have contributed to the lower rates of neuropsychiatric disorders. The social stigma around mental health may prevent the sharing of this information, and social and behavioural difficulties may lead to a loss of contact with family members. Furthermore 74/305 (24.26%) patients did not have any family history information. We may also have missed patient cases of neuropsychiatric disorders as we were unable to capture those with subclinical symptoms not meeting traditional diagnostic criteria, those without an official diagnosis, those who did not seek medical treatment and with previously under-recognised disorders (ie, autism spectrum disorders). Future research using more comprehensive and complete family and patient data may reveal stronger associations.

C9orf72 status was available for 259 (85%) patients in this study, allowing us to account for this in secondary analysis. Although we showed that almost all associations between patient and family neuropsychiatric history, and cognition and behaviour remained after controlling for the patients’ C9orf72 status, we did not control for any other factors. Future studies may also include other confounding variables and other genes or whole-genome sequencing previously associated with cognitive and behavioural impairment.41 Furthermore, future research should include other patient groups and healthy controls to determine whether these associations are specific to MND.

Finally, we focused on results unadjusted for multiple comparisons increasing the risk of a false-positive result. This study is experimental and hypothesis generating; therefore, it is important to show unadjusted effects as these warrant further examination in future research.43 However, unadjusted findings should be interpreted with caution.

Conclusion

This is the first study to show that neuropsychiatric disorders in patients (premorbidly) and their family members are associated with cognitive and behavioural changes in MND. These associations were evident in those with premorbid neuropsychiatric disorders, those without concurrent FTD and those who remained after accounting for C9orf72 status. These results provide further evidence of an overlap between MND, FTD and neuropsychiatric disorders and highlight the importance of further large-scale studies, using a more in-depth measurement of neuropsychiatric disorders and family history.

Acknowledgments

The authors thank the people with motor neuron disease (MND) who participated in this study, the MND research nurses, the Scottish

Clinical, Audit, Research and Evaluation of MND (CARE-MND) Consortium (Siddharthan Chandran, Robert Swingler, George Gorrie, Ian Morrison, Judith Newton, Suvankar Pal, Sharon Abrahams, Richard Davenport, Anthony Bateman, Colin Smith, Mary Porteous, Jon Warner, Elaine Cleary, Dorothy Storey, Moira Flett, Dianne Fraser, Susan Stewart, Andrew Bethell, Kitty Millar, Carolyn Webber, Gill Craig, Laura Marshall, Laura Cunningham, Suzanne Byrne, Janice Hatrick, Helen Lennox, Gill Stott and Alison McEleney) and the CARE-MND platform, hosted by the Euan MacDonald Centre for MND Research and funded by MND Scotland, and South East Scotland Genetics Service for C9rf72 screening.

References

View Abstract

Footnotes

  • Twitter @c_mchutchison

  • Contributors CAM contributed to the conception of the project, carried out the data analysis and drafted the manuscript. DJL collected the data included in the analysis and along with SC and SP was involved in the update of the CARE-MND platform. AM, SP and SA contributed to the conception of the project and provided input on the statistical analysis. EMC, JW and MP conducted the C9orf72 screening. The manuscript was reviewed by all authors.

  • Funding Centre for Cognitive Ageing and Cognitive Epidemiology, Department of Psychology, University of Edinburgh, UK.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval Ethical approval was obtained for the original (MREC/98/0/56 1989–2010, 10/MRE00/78 2011–2015) and updated CARE-MND platform (Scotland A Research Ethics Committee, 15/SS/0126).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available upon reasonable request.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles