A study examining the relationship between alexithymia and challenging behaviour in adults with intellectual disability.

bs_bs_banner

Journal of Intellectual Disability Research

doi: 10.1111/jir.12186

1022 volume 59 part 11 pp 1022 –1032

NOVEMBER

2015

A study examining the relationship between alexithymia and challenging behaviour in adults with intellectual disability B. E. Davies,1 N. Frude,2 R. Jenkins,2 C. Hill3 & C. Harding3 1 Abertawe Bro Morgannwg University Health Board, Psychology, Caswell Clinic, Glanrhyd Hospital, Bridgend, Wales, UK 2 Doctoral Course in Clinical Psychology, Cardiff University, Cardiff, Wales, UK 3 Aneurin Bevan Health Board, Learning Disability Services, Gwent, Wales, UK

Abstract Background Research suggesting that people with intellectual disabilities (ID) have difficulties in recognising emotions provides a rationale for studying alexithymia in this population. A number of studies have found a relationship between alexithymia and challenging behaviours in various populations and this study aims to discover if this is the case for people with ID. Method Cross-sectional data were collected from 96 participants with ID and 95 of their carers. The service user participants completed an alexithymia questionnaire for children while carers completed the checklist for challenging behaviour and the observer alexithymia scale. Correlational analyses were employed to explore relationships between the variables. Results The relationship between service user and carer-rated alexithymia was very weak. The analysis did show significant associations between observerrated alexithymia and challenging behaviour frequency, management difficulty and severity, but Correspondence: Dr Bronwen Elizabeth Davies, Abertawe Bromorgannwg University Health Board, Psychology, Caswell Clinic, Glanrhyd Hospital, Tondu Road, Bridgend, Wales CF31 4LN, UK (e-mail: [email protected]).

there was no significant relationship between challenging behaviour and alexithymia as rated by service users themselves. Conclusions This study suggests that observer-rated alexithymia is important in understanding challenging behaviour presented by people with ID. Service user-rated alexithymia had no association with challenging behaviour, in contrast to the results from similar research with other challenging populations. Keywords adult, alexithymia, challenging behaviour, emotional perception, intellectual disability

Introduction Alexithymia has been characterised as a deficit in emotional intelligence. It has been defined as the inability to differentiate, describe and label one’s own emotions and literally means ‘no words for feelings’ (Sifneos 1973; Bagby et al. 1994; Taylor et al. 1999). Features of alexithymia include difficulties in identifying and distinguishing between feelings and bodily sensations, difficulties in labelling and communicating emotional experience and externally oriented thinking (Taylor et al. 1999).

© 2015 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd


volume 59 part 11

Nov e mBer

2015

1023 B. E. Davies et al. • Alexithymia and challenging behaviour

Alexithymia has been found to be associated with substance misuse, various mental health difficulties (Fukunishi et al. 1999), physical ill health (Taylor & Bagby 2004) and personality disorder (Berenbaum 1996). Researchers have not yet explored alexithymia in people with intellectual disabilities (ID). Mellor & Dagnan (2005) argue that there is a strong case for research in this area because of parallels between the alexithymia construct and the emotional recognition difficulties and external cognitive styles that are characteristic of people with ID. They also point out that some of the factors known to be associated with the development of alexithymia (e.g. trauma and abuse, poor bonding and insecure attachment in infancy) are often present in the lives of people with ID (Mellor & Dagnan 2005). Researchers have found a negative correlation between alexithymia and verbal intelligence quotient (IQ) scores in offender populations. As people with lower IQs are more alexithymic, it may be relevant to study the degree and effects of alexithymia in people with ID (Kroner & Forth 1995; Louth et al. 1998). If one is unable to interpret emotional stimuli, socio-emotional development may be impeded and antisocial or challenging behaviour, withdrawal or mood disorders may emerge (Zaja & Rojahn 2008). Researchers argue that difficulties in attending to, identifying and communicating emotions place individuals at an increased risk of engaging in aggressive and self-harm behaviours as ways of expressing their emotional pain and distress (Paivio & McCulloch 2004). Studies have looked at the relationship between alexithymia and complex or challenging behaviours within various populations (e.g. Louth et al. 1998; Paivio & McCulloch 2004; Hornsveld & Kraaimaat 2012), but not to date with people with ID. Research studies have looked at the relationship between alexithymia and violence, the majority of these being conducted with offenders. Violent offenders present with significantly higher levels of alexithymia than non-violent control groups (Louth et al. 1998; Hornsveld & Kraaimaat 2012), with alexithymia being correlated with measures of anger, hostility and aggression (Hornsveld & Kraaimaat 2012). Similarly, with adolescents, high levels of alexithymia are associated with aggression

and offending behaviour (Zimmermann 2006; Manninen et al. 2011; Konrath et al. 2012). In nonoffender populations, Teten et al. (2008) studied a largely male veteran population and found alexithymia to be a significant predictor of impulsive aggression. Payer et al. (2011) studied methamphetamine abusers and found that difficulty identifying feelings (DIF) [from the Toronto Alexithymia Scale (TAS-20)] was related to self-reported aggression. Levels of alexithymia have also been shown to be high in people who harm themselves. This association has been found in a variety of populations, e.g. adolescents and students (Lambert & de Man 2007; Borrill et al. 2009; Garisch & Wilson 2010), female psychiatric in-patients (Zlotnick et al. 1996) and illicit substance users (Oyefeso et al. 2008). Researchers have also found the DIF factor on the TAS-20 to be a significant predictor of self-harm in adolescents (Lambert & de Man 2007; Borrill et al. 2009) and substance misuse populations (Oyefeso et al. 2008). Two studies failed to find significant relationships between alexithymia and challenging behaviour. Swannell et al. (2012) accessed 11 423 adults who had participated in the ‘Australian Epidemiological Study of Self-Injury’ and found that alexithymia was only a weak predictor of self-harm. However, they only measured the ‘difficulty describing feelings’ (DDF) factor of the TAS-20 in their assessment of alexithymia and this factor does not appear to have such a strong relationship with self-harm as the DIF factor (Borrill et al. 2009; Swannell et al. 2012). Moriarty et al. (2001) found no significant difference in alexithymia scores between male adolescent sex offenders and non-offenders, but the small sample size (n = 15) in this study might well mean that this finding is unreliable (Hornsveld & Kraaimaat 2012). In summary, there does appear to be a relationship between alexithymia and challenging behaviours in several populations, particularly in relation to the DIF factor of the TAS-20. Although there may be some overlap in terms of the behaviours presented by other populations, one needs to be cautious in generalising these findings to an ID community population for a number of reasons. Firstly, forensic samples largely use male participants and self-harm samples largely use female samples, limiting generalisability across genders.



volume 59 part 11

Nov e mBer

2015


The research discussed earlier has tended to use younger samples across a narrower age range than the ID research. Also, there is no evidence that the behaviours examined in these studies are similar in terms of frequency, management difficulty or severity, with most of them not including any measures of challenging behaviour. The challenging behaviours of antisocial or vulnerable populations are likely to be functionally different from those of people with ID, with differing intentions and consequences. All these factors limit the generalisability of these findings to an ID population. These studies are, however, relevant in establishing a link between alexithymia and challenging behaviour and commonalities do exist. For example, low IQs are often found in offender populations (Prison Reform Trust 2007). This indicates the need for research to identify whether alexithymia is a relevant factor in the challenging behaviour presented by people with ID. The hypothesis for this study is therefore that alexithymia scores will be positively correlated with the frequency, management difficulty and severity of challenging behaviour in an ID population. This study will also consider whether selfrated and carer-rated alexithymia are related and which is the most useful predictor of challenging behaviour.

Method Design The study used a correlational design to explore relationships between the variables of self-rated and carer-rated alexithymia and carer-rated levels of challenging behaviours in adults with an ID. The questionnaire method of data collection was employed with both service users and carers. The dependent variables were the frequency, management difficulty and severity of challenging behaviours including ‘aggressive’ and ‘other challenging behaviours’. The independent variable was alexithymia.

Sample Although no previous studies have examined the relationship between the key variables examined in

this study within a population of people with ID, a number of studies can be seen as ‘near neighbours’. Thus, Zlotnick et al. (1996) found a correlation of 0.33 between alexithymia and deliberate self-harm in a population of 153 female psychiatric inpatients; Lambert & de Man (2007) found a correlation of 0.39 between self-mutilation and alexithymia in a population of French adolescent girls who were accessing psychological services and Teten et al. (2008) found a correlation of 0.32 between alexithymia and impulsive aggression in a population of 82 male veterans. Using the results of these studies as possible indicators of the likely strength of associations between the variables, with a medium effect size correlation of 0.33, and with 0.8 power and probability level 0.05 (and one-tail hypotheses), the sample size needed was identified as 55 participants (55 service users together with their 55 carers) to provide sufficient power to identify the relevant effects. As the prevalence of challenging behaviour in the population of people with ID is approximately 10% (Emerson et al. 2001; Lowe et al. 2007), it was decided to recruit a sample that included an enhanced number of people presenting with challenging behaviours. Participants were recruited from two south Wales ID services, through community teams and specialised challenging behaviour services. To be included within the study, participants needed to have a mild or moderate ID, to be over 18, to have an ability to communicate verbally and to have the capacity to consent to participation. Those approached to participate in this research were identified as having a mild to moderate ID by their clinical team. No formal assessment was completed. The only exclusion criterion was a diagnosis of an autistic spectrum condition, as it is known that people with these conditions have difficulty recognising and processing the emotional states of others and responding in emotionally appropriate ways (Owen et al. 2001). In total, 116 potential participants were identified by the teams. Of these, 16 declined to participate and four lacked the capacity to consent to participation or lacked the ability required to complete the questionnaires and were therefore excluded from the study. Carers were recruited after being identified by service users as someone who knew them well and for at least 6 months. Thus, the study



volume 59 part 11

Nov e mBer

2015


sample consists of 96 people with ID. Carer participants supplied the data for all but one of these participants.

of this questionnaire. Participants selected their response using a visual scoring system.

The observational alexithymia scale Measures The alexithymia questionnaire for children The TAS-20 (Bagby et al. 1994) is the most frequently used measure of alexithymia within the literature. The ‘alexithymia questionnaire for children’ (AQC) (Rieffe et al. 2006) is a simplified version of the TAS-20, enabling its use with primary schoolaged children. The AQC measures three factors that represent core features of alexithymia: DIF, DDF and ‘externally oriented thinking’ (EOT). It employs a 3-point response scale for each item (0 = not true, 1 = sometimes true, 2 = often true). The AQC was considered more appropriate for this study as the items are simpler and shorter than those from the TAS-20. Although it was designed for use with children, the AQC does not make reference to any specifically child-related concepts such as ‘play’ or ‘toys’. Piloting the instrument with adults with ID, however, did suggest the need for further adaptations because some of the questions appeared to prompt participants towards a particular answer and some of the language used appeared difficult to comprehend and needed further simplification. The AQC was validated with a non-intellectually disabled child population (ages 9 to 15) (Rieffe et al. 2006). Two of the factors, DIF and DDF, showed good predictive validity and were significantly positively correlated with a somatic checklist and negative emotion mood scales. These two factors also showed good internal consistency (Cronbach’s alpha around 0.75 for each). Within the ID population within this study, Cronbach’s alpha was calculated and showed internal consistency of 0.43 for the DDF factor and 0.64 for the DIF factor. The EOT factor failed to meet the criteria for internal consistency or predictive validity in Rieffe et al.’s (2006) paper; this was also the case within the current ID population. The data relating to the EOT factor is therefore excluded from this study. This is consistent with research findings relating to the TAS-20 (Kooiman et al. 2002). To ensure consistency, a protocol was developed for the delivery

The 33-item observational alexithymia scale (OAS) (Haviland et al. 2000) was designed as an instrument that can be completed by service users’ relatives or acquaintances to measure alexithymia. Each item is rated on a 4-point scale ranging from ‘0 – Never: Not at all like this person’ to ‘3 – All the time: Completely like this person’. The scale includes 18 positively scored items and 15 reverse scored items. The instrument measures five factorial domains, these being: ‘distant’, ‘uninsightful’, ‘somatising’, ‘humourless’ and ‘rigid’. Confirmatory factor analyses showed strong correlations between the first-order factors (‘distant’, ‘uninsightful’, ‘somatising’, ‘humourless’ and ‘rigid’) and the second-order construct of alexithymia (P < 0.05). In terms of reliability, the test–retest coefficient was 0.87, showing that OAS scores remained relatively stable over a 2-week period. The authors argue that the internal consistency, stability and factorial invariance provide support for the OAS’ construct validity. Cronbach’s alpha calculated within this carer population ranged between 0.52 for ‘rigid’ and 0.83 for ‘distant’, indicating acceptable levels of internal consistency.

The checklist of challenging behaviour The checklist of challenging behaviour (CBC) was completed by carers (Harris et al. 1994). The measure contains two checklists of behaviours, the first one listing 14 ‘aggressive behaviours’ and the second one listing 18 ‘other challenging behaviours’. Scales are used to rate the behaviours in terms of ‘frequency’, ‘management difficulty’ and, for the aggressive behaviours only, ‘severity’. The ‘frequency’ scale ranges from ‘0 – Never shown this behaviour to my knowledge’ to ‘6 – Very frequently – Daily or more often in the past month’. The ‘management difficulty’ scale assesses the rater’s perception of their own difficulty managing a challenging situation. This ranges from ‘0 – No problem – I can usually manage this situation with no difficulty’ to ‘4 – Extreme problem – I simply cannot manage this situation without help’. The ‘severity’


volume 59 part 11


Nov e mBer

2015


scale focuses on the degree of tissue damage and ranges from ‘0 – No injury – Does not appear to cause pain or tissue damage to other person’ to ‘4 – Very serious injury – Has caused very serious tissue damage (e.g. broken bones, deep lacerations/ wounds) or resulted in hospitalisation and/or certified absences from work for whatever reasons’. Carers were asked to rate the frequency of each behaviour. If they rated it ‘0’, then they could move on to the next behaviour, but if they recorded any other number in the frequency box they were asked to rate the management difficulty and severity for that behaviour. Within this study, an overall score was calculated for the ‘frequency of challenging behaviour’, ‘management difficulty of challenging behaviour’ and ‘aggression severity’ and these composite scores were used in the data analyses. Harris et al. (1994) assessed the CBC for interrater reliability, between interviewer reliability and test–retest reliability. The results showed acceptable levels of reliability, with critical values of rs being significant at the P < 0.05 level for all three rating scales (frequency, management difficulty and severity). Harris et al. (1994) did, however, note a tendency for the reliability of the scales to decrease as the numbers of the behaviours recorded increased. In terms of validity, the items that were included were based on information from service providers, a review of other checklists for challenging behaviours and an examination of hospital records of violent incidents. The results of this indicated that the content validity of the CBC is high.

information was provided to the service user and, if they were happy to proceed, they were asked to sign a consent form. The service user was asked to identify a carer who the researcher could approach. This had to be someone who had known them for at least 6 months. Additional information sheet and consent form were provided to the carer who, with their agreement, was asked to complete the CBC and the OAS. The order of the questionnaire completion was random and the choice of carer participants. The participants with a learning disability were asked to complete the adapted AQC supported by the researcher reading out the questions and prompting them to select their answer from visual stimuli placed in front of them. Where possible, carers were asked to complete questionnaires at the same time but, alternatively, this could be done via e-mail or post. All the data were collected by the first author and a psychology placement student who had been trained to administer the questionnaires by the first author. Participants were provided with the researcher’s contact details should they have any questions or wished to make a complaint. They were also offered the opportunity to obtain feedback on the outcomes of the research when the project was complete. The questionnaires were anonymous and data generated from each participant was not accessible to anyone other than the researcher and the academic and clinical supervisors. Data were analysed using the Statistical Package for Social Sciences version 20 (IBM Corporation, New York, United states of America).

Procedure Ethical permission to conduct this study was obtained from the National Health Service South West Wales Local Research Ethics Committee and the study was also approved by the Research and Development Departments of the relevant health boards. Information about the research project was presented at multidisciplinary team meetings by the first author. Potential participants were then approached by known professionals or administrative staff with information about the research. If service users agreed, their details were forwarded to the researcher and an arrangement was made to meet them. At this meeting further, accessible,

Results Participants Of the 96 service user participants, 46 were women and 50 were men. The age range of the participants was 18–79 years old, with the mean age of the sample being 39.68 years, the median 39.0 years and the standard deviation 13.32 years. Of the 95 carer participants, 72 were women and 19 were men, and information on gender was missing for the four other carers. The carers who were nominated by the service users had known them between 6 months and 54 years and represented a number of caring roles. As would be expected, the carers who


volume 59 part 11


Nov e mBer

2015


had known service users longest were generally family members. Carers had known their service users for an average of 7.94 years (this was the mean; the median was 5 years). The majority (80%) of carers had known the service users less than 10 years and 57% had known them for less than 5 years. The roles of carer participants and the time they had been known to the service users are reported in Table 1.

Descriptive statistics Descriptive statistics relating to the variables measured in the study are presented in Table 2.

Examination of the distribution of scores established that the data relating to each of the AQC scales, including the total score, did not depart significantly from a normal distribution. Similarly, the OAS total score and three of the sub-scales on this measure were normally distributed (‘humourless’, ‘distant’ and ‘uninsightful’). When a pair of variables was both normally distributed, a Pearson’s correlation was used to assess the degree of association. When one of both of a pair was not normally distributed (it was established that this was the case for the ‘somatising’ and ‘rigid’ scales of the OAS), correlations were computed using Spearman’s rho. Following these methods, very little relationship was

Table 1 The relationship between carer participants and service users and the average time known to one another

Time known to service user (years) Carer’s role

n

Mean

Median

Range

Support worker Intellectual Disability service professional (nurse, social worker etc.). Home manager/team leader Family member (6 mothers, 1 partner) Key worker Day service staff (4 missing values) Adult placement carer Total n = 91 (4 missing values)

30 20 22 7 5 9 2 95

5.15 5.45 4.98 30.13 3.90 7.40 30.00 7.75

5.00 4.00 4.00 32.50 3.50 6.00 30.00 5.00

1–18 0.5–20 0.5–13 8–45 1–7 4–14 6–54 0.5–54

Table 2 Table showing descriptive statistics for the challenging behaviour and alexithymia scales

Checklist for challenging behaviour Challenging behaviour frequency Challenging behaviour management difficulty Aggression severity Alexithymia questionnaire for children (AQC) Difficulty identifying feelings Difficulty describing feelings Total AQC Observer alexithymia scale (OAS) Rigid Humourless Somatising Uninsightful Distant Total OAS

n

Range

Mean

Median

Standard deviation

94 91 94

0–83 0–59 0–19

19.19 8.46 2.31

11.5 2 0

20.19 12.35 4.52

96 96 96

0–14 1–10 8–31

6.27 5.56 20.40

6 6 20.5

3.13 2.22 4.96

92 94 94 92 92 90

0–12 0–13 0–14 3–22 2–28 16–68

4.95 4.76 4.77 12.63 14.45 41.44

4 4 5 12 14 42

2.78 2.88 3.19 4.51 5.43 1.33


volume 59 part 11


Nov e mBer

2015


Table 3 Correlations between alexithymia and challenging behaviour

Alexithymia ‘difficulty identifying feelings’ Alexithymia ‘difficulty describing feelings’ Alexithymia total Observer alexithymia distant Observer alexithymia uninsightful Observer alexithymia somatising Observer alexithymia humourless Observer alexithymia rigid Observer alexithymia total

Challenging behaviour frequency

Challenging behavior management difficulty

Aggression severity

0.079 0.160 0.150 0.218* 0.491** 0.154 0.194 0.320** 0.497**

0.059 0.126 0.133 0.170 0.325** 0.067 0.124 0.084 0.317**

–0.059 0.186 0.133 0.147 0.296** 0.133 0.096 0.150 0.298**

* P < 0.05, ** P < 0.01.

found between scores on the OAS and the AQC. The only significant positive correlations found, using a two-tailed Spearman’s rho test, was between ‘somatising’ (OAS) and ‘AQC total’ (rs = 0.250, P < 0.05). The frequency distributions were examined by calculating the z-score for skewness and kurtosis by dividing each value by its standard error. A z-score of 1.96 or above indicates an unsatisfactory level of skewness or kurtosis for parametric analyses as it implies that the data are not normally distributed. The challenging behaviour variables were found not to be normally distributed and therefore nonparametric correlational analyses were performed. Two-tailed Spearman’s rho correlations were computed to measure the strength of association between challenging behaviour and alexithymia. It was thought that two-tailed correlations would provide both a more conservative and robust analysis because of the number of correlations computed. No correlations reached significance between service user measured alexithymia (AQC) and challenging behaviour. Significant positive correlations were, however, found between carer-rated alexithymia (OAS) and some of the challenging behaviour scores. Challenging behaviour ‘frequency’ was significantly related to the sub-scales ‘distant’ (rs = 0.218, P < 0.05), ‘uninsightful’ (rs = 0.491, P < 0.01), ‘rigid’ (rs = 0.320, P < 0.01) and ‘observer alexithymia total’ (rs = 0.497, P < 0.01). Challenging behaviour ‘management difficulty’ was significantly correlated

with ‘uninsightful’ (rs = 0.325, P < 0.01) and ‘observer alexithymia total’ (rs = 0.317, P < 0.01). Finally, aggression ‘severity’ was significantly correlated with ‘uninsightful’ (rs = 0.296, P < 0.01) and ‘observer alexithymia total’ (rs = 0.298, P < 0.01). Correlations between challenging behaviour and alexithymia are shown in Table 3.

Discussion The only significant relationships that were found were between the OAS and the AQC were ‘somatising’ (OAS) and ‘total AQC’. In previous studies, the OAS and the TAS-20 (which the AQC is based on) did not correlate very highly and it has been suggested that they may relate to different aspects of alexithymia (Lumley et al. 2005; Meganck et al. 2010). An observer can clearly not comment on the inner world of an individual and therefore the OAS looks at the behavioural presentation that would be expected if someone was alexithymic. The limited relationship between these scales may be partially because of difficulties with the conceptualisation of alexithymia. The AQC and TAS-20 only measure the awareness and verbalisation of one’s feelings, which may not represent the breadth of alexithymia (Kooiman et al. 2002; Lumley et al. 2005). Alexithymia is sometimes defined as a global impairment in the processing of emotion and includes features such as somatisation and lack of humour, as in the OAS



volume 59 part 11

Nov e mBer

2015


(Lumley et al. 2005). Meganck et al. (2010), however, argue that the broader definition of alexithymia goes beyond the concept itself and includes elements that may be related to alexithymia but are not intrinsic to it. They believe that variables such as somatisation, rigidity and distance may correlate with alexithymia but should not be regarded as facets of the condition. From this, it is argued that a high score on the OAS may not be a pure measure of alexithymia. Difficulties with the definition of alexithymia may also contribute to problems in finding a clear factor structure in both the TAS-20 (and AQC) and the OAS (Meganck et al. 2010). Lumley et al. (2005) argue that until the field has greater theoretical clarity, researchers need to use multiple measures. The hypothesis that alexithymia scores would be positively correlated with the frequency, management difficulty and severity of challenging behaviour was partially supported by the results. No relationship was found between service user measured alexithymia and challenging behaviour. Scores on carer-rated alexithymia (OAS), however, were related to all aspects of challenging behaviour. Challenging behaviour ‘frequency’ was significantly related to the OAS ‘total’ and to the sub-scales ‘distant’, ‘uninsightful’, ‘rigid’ and ‘humourless’. Challenging behaviour ‘management difficulty’ was significantly correlated with OAS ‘total’ and ‘uninsightful’. Finally, ‘aggression severity’ was significantly correlated with OAS ‘total’ and ‘uninsightful’. Two-tailed correlations were used to ensure a more conservative and robust approach, however, because of the number of correlations computed, it is possible that some may have occurred by chance, and one therefore has to be cautious in interpreting these results, particularly in relation to those that only reached 0.05 level of significance. This is the first study to examine how alexithymia relates to challenging behaviour presented by people with ID. The findings of this study do support previous literature that shows a relationship between alexithymia and challenging behaviour in other populations (e.g. Teten et al. 2008; Konrath et al. 2012), but one must be cautious in comparing this study with previous studies. Earlier research used self-reported measures of alexithymia which, in this study, failed to yield a significant relationship with

challenging behaviour. Further studies of the relationship between observer-rated alexithymia (OAS) and challenging behaviour need to be conducted with other populations in order to reach a better understanding of this relationship. There are a number of limitations and to assume that alexithymia may lead to challenging behaviour would be to ignore the possible effects of confounding variables including IQ and language ability which were not measured in this study. Selfreported alexithymia may be confounded by level of ID; previous research has shown significant relationships between emotional recognition ability and intelligence or language ability (Simon et al. 1995, 1996; McEvoy et al. 2002; Joyce et al. 2006). In this study, a measurement was not taken and participants were identified by clinicians in their team as having a mild or moderate ID, this may therefore be a confounding variable that needs further examination in future research. This would have been more of an issue if there were any relationships observed between service user measured alexithymia and challenging behaviour. This was not, however, the case and the significant findings relating alexithymia to challenging behaviour were all based on carerobserved alexithymia. Future research could examine the relationship between intelligence (IQ) or language ability and self-reported alexithymia in people with ID. In this study, carers were asked to complete the two questionnaires and did this in a random order, it is possible that if they chose to complete the CBC first this may have influenced their responses in the OAS, causing them to focus on the negative aspects of the service user and introducing bias. Higher correlations were found between OAS scores and challenging behaviour, both of these measures being completed by carers. This may be evidence of a ‘halo error’, the error that occurs when the rater’s overall impression or evaluation of a person strongly influences their ratings, inflating or creating illusory intercorrelations between the factors measured (Lance et al. 1990; Murphy et al. 1993; Solomonson & Lance 1997). It is possible that carers rated service-user participants, in this study, more negatively with regard to alexithymia because of their view of that person, in light of their challenging behaviour. There is some debate, however, about whether the halo error negatively impacts on accu-


volume 59 part 11


Nov e mBer

2015


racy. Fisicaro (1988) found a negative relationship between halo error and accuracy, although some studies have found the contrary leading Murphy et al. (1993), in their review, to conclude that the halo error does not necessarily imply low levels of accuracy and may in fact increase the accuracy and validity of ratings. They also argue that efforts to control the halo error have not proved successful. Certain factors are recognised as limiting the halo effect. For example, the more familiar the rater is with the ratee and the more dimensions they are asked to rate on, the lesser the halo error. These factors may have reduced the halo effect in this study. Murphy et al. (1993) point out that it is often difficult to determine whether halo errors have occurred or what to do about them. There has been significant debate over the validity of using self-report in measuring alexithymia. Alexithymic individuals, by their nature, are not very self-reflective and are likely to have deficient or impaired introspection. Using self-report measures therefore requires an alexithymic individual to report on a capacity they lack (Kooiman et al. 2002; Zimmermann 2006; Meganck et al. 2010). Unlike previous research, service user-measured alexithymia (AQC) was not related to challenging behaviour. The abstract nature of the questions may have caused confusion to participants, being too complex for people with ID who are likely to cope better with more concrete questions (Lynch 2004). These results also have a number of clinical implications, relevant within both therapy and home environment. Evidence suggests that one can help individuals to become more emotionally aware and expressive in a safe, supportive and empathic setting (Zimmermann 2006). Staff teams should be supported to develop an understanding of the link between emotional perception and regulation difficulties and challenging behaviour. This would be an opportunity for them to develop skills to enhance the emotional perception abilities of the service users they support. For example, it can be helpful for carers or therapists to offer emotional labels for service users’ current and past experiences and to identify previously unrecognised triggers for emotion, facilitating emotional awareness and communication (Ogrodniczuk et al. 2005). In addition, group therapy, group training programmes around emotions and cognitive behaviour therapy have had

positive outcomes for those with difficulties understanding their emotions (Wood & Stenfert-Kroese 2007; Spek et al. 2008; Ogrodniczuk et al. 2012). Some studies suggest that therapeutic outcomes are compromised for people with alexithymia because of difficulties in establishment and maintenance of a therapeutic relationship (Ogrodniczuk et al. 2005, 2012). Ogrodniczuk et al. (2005) found that therapists reacted more negatively to people with high levels of alexithymia, viewing them as having fewer positive qualities, being less compatible with them and having little significance as members of the group in therapy. This led to poorer outcomes for people with alexithymia who may have experienced a lack of support, belonging and mutual understanding. This reaction may also be experienced by staff teams working with individuals presenting challenging behaviours who similarly have high levels of alexithymia. Such countertransference reactions should be addressed within supervision and may help carers or therapists to gain insight into the service user’s inner life and promote empathy and understanding (Ogrodniczuk et al. 2005). In conclusion, this study identifies a significant relationship between observer-rated alexithymia and challenging behaviour in people with ID. It has been suggested that there is a strong case for research in this area because of parallels between the alexithymia construct and the emotional recognition difficulties and external cognitive styles that have been identified in people with ID (Mellor & Dagnan 2005). In this study, service user-rated alexithymia had no significant relationships with challenging behaviour, in contrast to research with other challenging populations. A number of limitations have been outlined and suggestions for further research have been made.

References Bagby R., Parker J. & Taylor G. (1994) The twenty-item Toronto Alexithymia Scale-I. item selection and cross validation of the factor structure. Journal of Psychosomatic Research 38, 23–32. Berenbaum H. (1996) Childhood abuse, alexithymia and personality disorder. Journal of Psychosomatic Research 41, 585–95.



volume 59 part 11

Nov e mBer

2015


Borrill J., Fox P., Flynn M. & Roger D. (2009) Students who self-harm: coping style, rumination and alexithymia. Counselling Psychology Quarterly 22, 361–72. Emerson E., Kiernan C., Alborz A., Reeves D., Mason H., Swarbrick R. et al. (2001) The prevalence of challenging behaviours: a total population study. Research in Developmental Disabilities 22, 77–93. Fisicaro S. (1988) A re-examination of the relation between halo error and accuracy. Journal of Applied Psychology 73, 239–44. Fukunishi I., Sei H., Morita Y. & Rahe R. (1999) Sympathetic activity in alexithymics with mothers low care. Journal of Psychosomatic Research 46, 579–89. Garisch J. & Wilson M. (2010) Vulnerabilities to deliberate self- harm among adolescents: the role of alexithymia and victimization. British Journal of Clinical Psychology 49, 151–62. Harris P., Humphreys J. & Thomson G. (1994) A checklist for challenging behaviour: the development of a survey instrument. Mental Handicap Research 7, 118–33. Haviland M., Warren W. & Riggs M. (2000) An observer scale to measure alexithymia. Psychosomatics 41, 385–92. Hornsveld R. & Kraaimaat F. (2012) Alexithymia in Dutch violent forensic psychiatric out-patients. Psychology, Crime & Law 18, 833–46. Joyce T., Globe A. & Moody C. (2006) Assessment of the component skills for cognitive therapy in adults with intellectual disability. Journal of Applied Research in Intellectual Disabilities 19, 17–23. Konrath S., Novin S. & Li T. (2012) Is the relationship between alexithymia and aggression context-dependent? Impact of group membership and belief similarity. Personality and Individual Differences 53, 329–34. Kooiman C., Spinhoven P. & Trijsburg R. (2002) The assessment of alexithymia. A critical overview of the literature and a psychometric study of the Toronto Alexithymia Scale-20. Journal of Psychosomatic Research 53, 1083–90. Kroner D. & Forth A. (1995) The Toronto Alexithymia Scale with incarcerated offenders. Personality and Individual Differences 19, 625–34. Lambert A. & de Man A. (2007) Alexithymia, depression and self mutilation in adolescent girls. North American Journal of Psychology 9, 555–66.

Lumley M., Gustavson B., Partrige T. & Labouvie-Vief G. (2005) Assessing alexithymia and related emotional ability constructs using multiple methods: interrelationships among measures. Emotion (Washington, D.C.) 5, 329–42. Lynch C. (2004) Psychotherapy for persons with mental retardation. Mental Retardation 42, 399–405. McEvoy J., Reid Y. & Guerin S. (2002) Emotion recognition and concept of death in people with learning disabilities. The British Journal of Developmental Disabilities 48, 83–9. Manninen M., Therman S., Suvisaari J., Ebeling H., Moilanen I., Huttman M. et al. (2011) Alexithymia is common among adolescents with severe disruptive behaviour. The Journal of Nervous and Mental Disease 199, 506–9. Meganck R., Vanheule S., Desmet M. & Inslegers R. (2010) The observer alexithymia scale: a reliable and valid alternative for alexithymia measurement? Journal of Personality Assessment 92, 175–85. Mellor K. & Dagnan D. (2005) Exploring the concept of alexithymia in the lives of people with learning disabilities. Journal of Intellectual Disabilities 9, 229–39. Moriarty N., Stough C., Tidmarsh P., Eger D. & Dennison S. (2001) Deficits in emotional intelligence underlying adolescent sex offending. Journal of Adolescence 24, 743–51. Murphy K., Jako R. & Anhalt R. (1993) Nature and consequences of halo error: a critical analysis. Journal of Applied Psychology 78, 218–25. Ogrodniczuk J., Piper W. & Joyce A. (2005) The negative effect of alexithymia on the outcome of group therapy for complicated grief: what role might the therapist play? Comprehensive Psychiatry 46, 206–13. Ogrodniczuk J., Sochting I., Piper W. & Joyce A. (2012) A naturalistic study of alexithymia among psychiatric outpatients treated in an integrated group therapy program. Psychology and Psychotherapy: Theory, Research and Practice 85, 278–91. Owen A., Browning M. & Jones R. (2001) Emotion recognition in adults with mild-moderate learning disabilities. Journal of Intellectual Disabilities 5, 267–81.

Lance C., Fisicaro S. & LaPointe J. (1990) An examination of negative halo error in ratings. Educational and Psychological Measurement 50, 545–54.

Oyefeso A., Brown S., Chiang Y. & Clancy C. (2008) Self- injurious behaviour, life events and alexithymia among treatment- seeking opiate addicts: prevalence, pattern and correlates. Drug and Alcohol Dependence 98, 227–34.

Louth S., Hare R. & Linden W. (1998) Psychopathy and alexithymia in female offenders. Canadian Journal of Behavioural Science 30, 91–8.

Paivio S. & McCulloch C. (2004) Alexithymia as a mediator between childhood trauma and self- injurious behaviours. Child Abuse and Neglect 28, 339–54.

Lowe K., Allen D., Jones E., Brophy S., Moore K. & James W. (2007) Challenging behaviours: prevalence and topographies. Journal of Intellectual Disability Research 51, 625–36.

Payer D., Lieberman M. & London E. (2011) Neural correlates of affect processing and aggression in methamphetamine dependence. Archives of General Psychiatry 68, 271–82.



volume 59 part 11

Nov e mBer

2015


Prison Reform Trust (2007) No One Knows: Offenders with Learning Difficulties and Learning Disabilities. The prevalence and Associated Needs of Offenders with Learning Difficulties and Learning Disabilities. Author, London. Rieffe C., Oosterveld P. & Meerum Terwogt M. (2006) An alexithymia questionnaire for children: factorial and concurrent validation results. Personality and Individual Differences 40, 123–33. Sifneos P. (1973) The prevalence of ‘alexithymia’ characteristics in psychosomatic patients. Psychotherapy and Psychosomatics 22, 255–62. Simon E., Rosen M., Grossman E. & Pratowski E. (1995) The relationships among facial recognition, social skills, and quality of life. Research in Developmental Disabilities 16, 383–91. Simon E., Rosen M. & Ponpipom A. (1996) Age and IQ as predictors of emotion identification in adults with mental retardation. Research in Developmental Disabilities 17, 383–9.

non-suicidal self- injury and the mediating roles of dissociation, alexithymia and self- blame. Child Abuse & Neglect 36, 572–84. Taylor G. & Bagby R. (2004) New trends in alexithymia research. Psychotherapy and Psychosomatics 73, 68–77. Taylor G., Bagby R. & Parker J. (1999) Disorders of Affect Regulation: Alexithymia in Medical and Psychiatric Illness. Cambridge University Press, Cambridge. Teten A., Miller L., Bailey S., Dunn N. & Kent T. (2008) Empathic deficits and alexithymia in trauma related impulsive aggression. Behavioural Sciences and the Law 26, 823–32. Wood P. & Stenfert-Kroese B. (2007) Enhancing the emotion recognition skills of individuals with learning disabilities: a review of the literature. Journal of Applied Research in Intellectual Disabilities 20, 576–9. Zaja R. & Rojahn J. (2008) Facial emotion recognition in intellectual disabilities. Current Opinion in Psychiatry 21, 441–4.

Solomonson A. & Lance C. (1997) Examination of the relationship between true halo and halo error in performance ratings. Journal of Applied Psychology 82, 665–74.

Zimmermann G. (2006) Delinquency in male adolescents: the role of alexithymia and family structure. Journal of Adolescence 29, 321–32.

Spek V., Nyklicek I., Cuijpers P. & Pop V. (2008) Alexithymia and cognitive behavior therapy outcome for subthreshold depression. Acta Psychiatrica Scandinavica 118, 164–7.

Zlotnick C., Shea T., Pearlstein T., Simpson E., Costello E. & Begin A. (1996) The relationship between dissociative symptoms, alexithymia, impulsivity, sexual abuse and self mutilation. Comprehensive Psychiatry 37, 12–16.

Swannell S., Martin G., Page A., Hasking P., Hazell P., Taylor A. et al. (2012) Child maltreatment, subsequent

Accepted 7 January 2015.


A preliminary study into the relationship between emotional perception ability and challenging behaviour in adults with an intellectual disability.

The relationship between emotional recognition ability and challenging behaviour in adults with an intellectual disability: a systematic review.

Assessment and treatment of challenging behaviour for individuals with intellectual disability: a research review.

Modification of motivational interviewing for use with people with mild intellectual disability and challenging behaviour.

Use of medication for challenging behaviour in people with intellectual disability.

Evaluating staff training: taking account of interactions between staff and clients with intellectual disability and challenging behaviour.

Clinical and cost effectiveness of staff training in Positive Behaviour Support (PBS) for treating challenging behaviour in adults with intellectual disability: a cluster randomised controlled trial.

pleasure in adults with severe and profound intellectual disabilities.

Central nervous system medication use in older adults with intellectual disability: Results from the successful ageing in intellectual disability study.

The relationship between the behavior problems and motor skills of students with intellectual disability.

Sedentary and Physical Activity Patterns in Adults with Intellectual Disability.

Obesity and associated factors in adults with intellectual disability.

The relationship between area deprivation and contact with community intellectual disability psychiatry.

A preliminary controlled trial of a trans-diagnostic programme for cognitive behaviour therapy with adults with intellectual disability.

A prospective study of the relationship between adverse life events and trauma in adults with mild to moderate intellectual disabilities.

The self-rated health of British adults with intellectual disability.

A register study of life events in young adults born to mothers with mild intellectual disability.

The relationship between skill deficits and disability-a transdiagnostic study.

Hospitalizations of adults with intellectual disability in academic medical centers.

Prevention of falls for adults with intellectual disability (PROFAID): a feasibility study.

Cognitive Behavioral Therapy for Depressed Adults with Mild Intellectual Disability: A Pilot Study.

Examining the relationship between subjective and objective memory performance in older adults: a meta-analysis.

The physical health of British adults with intellectual disability: cross sectional study.

Predictors of specialized inpatient admissions for adults with intellectual disability.