An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
National Guideline Alliance (UK). Mental Health Problems in People with Learning Disabilities: Prevention, Assessment and Management. London: National Institute for Health and Care Excellence (NICE); 2016 Sep. (NICE Guideline, No. 54.)
People with learning disabilities are more likely than people in the general population to experience living circumstances and life events associated with an increased risk of mental health problems, including birth trauma, stressful family circumstances, unemployment, deprivation, stigmatisation, lack of self-determination, and a lack of supportive friendships and intimate relationships (Cooper et al., 2007f; Martorell et al., 2009). In addition, people in this population are likely to have fewer psychological resources available to cope effectively with stressful events, including poorer problem-solving and planning skills associated with their cognitive impairments (van den Hout, 2000).
Over the last 30 years psychological therapies, especially cognitive behavioural therapy (CBT), have become established in the treatment of common mental health problems and some severe mental health problems such as psychosis (Department of Health, 2011). Many therapists have been reluctant, however, to offer these therapies to people with intellectual disabilities (Stenfert Kroese, 1998) because of the different challenges that come with developing a collaborative working relationship, and the difficulties of achieving treatment gains. Bender (1993) used the term the ‘the unoffered chair’ to describe this ‘therapeutic disdain’.
Recently there has been more professional interest in and research on the application of psychological therapies with people with learning disabilities (Taylor, 2013), but the supporting research is limited compared with that on the general population (Gustafsson et al., 2009; Prout & Browning, 2011). The most researched area with this population is that of anger associated with aggression and violence (Nicoll et al., 2013; Taylor & Novaco, 2013). This topic was covered in the NICE guidance on challenging behaviour and learning disabilities, which recommends that cognitive behavioural interventions are considered for adults with anger management problems (NICE, 2015; p255).
In relation to other mental health problems, effective psychological interventions for people with learning disabilities are based on careful assessment and individual formulation, as they are for other patient groups (Hatton, 2010). Beyond this, therapists will need to consider appropriate adaptations to the therapy environment and treatment framework, to enable people with learning disabilities to access and benefit from psychological therapies (Dagnan et al., 2013). In addition, modifications to the treatment interventions (manuals and protocols) will be required for clients with learning disabilities, and these adjustments will depend on the nature and degree of the impairments associated with each person’s learning disabilities (Lindsay et al., 2013). Therapists should adopt a functional approach to the adaptations and modifications that are required to enable people with learning disabilities to engage effectively in psychological therapy rather than basing judgements about an individual’s suitability for a particular intervention on their IQ measurement or global level of functioning (for example, milder learning disabilities).
In this chapter, recommendations are given for adaptations and modifications to psychological interventions that may be required for people with learning disabilities. Further recommendations are made for adapted interventions that should be considered for particular mental health problems experienced by people with learning disabilities based on the available research evidence.
The review protocol summaries, including the review questions and the eligibility criteria used for this section of the guideline, can be found in Table 22 and Table 23. A complete list of review questions and review protocols can be found in Appendix F; further information about the search strategy can be found in Appendix H.
Clinical review protocol summary for the review on the prevention of mental health problems.
Clinical review protocol summary for the review on the treatment and management of mental health problems.
As a result of limited quality evidence (section 5.2.2 below), the group decided to develop a set of general principles for adapting psychological treatments developed for people without learning disabilities so that such interventions can be delivered to people with learning disabilities. They developed these recommendations using the modified nominal group technique.
The modified nominal group technique used in this guideline is described in Chapter 3.
Key issues relating to the types of adaptations required when conducting psychological interventions to treat mental health problems in people with learning disabilities were identified by reviewing the available evidence (for example Hassiotis et al., 2013; Lindsay et al., 2015) and through GC meeting discussions. These sources were used to generate nominal statements to be rated by the GC. As the GC agreed that they did not have the sufficient expertise in psychological interventions, they identified practicing expert clinical psychologists to act as expert advisors. The expert advisors were asked initially simply to review the draft nominal statements. Following feedback from the advisors and consequent amendment, the nominal statements were distributed to the GC in the form of a questionnaire (round 1), for rating and comment.
Percentage agreement was calculated for each statement and comments were collated. The results were then presented and discussed. On initial review, the GC decided that the existing statements were not comprehensive and as not all expert advisers had responded on the first statements, the set of statements was re-generated using the comments and discussions of the GC members, as well as input from the advisors. These were again distributed in questionnaire format (round 2), for rating and comment. As the statements applied to interventions for people with milder learning disabilities, a separate set of statements were adapted for people with more severe learning disabilities, using the comments and discussions of the GC members, as well as input from the advisors.
As before, percentage agreement was calculated for each statement and comments collated. An example of a statement that was rated highly by the committee is: ‘The choice of intervention and introduction of adaptations should be informed by the person’s strengths and needs identified during assessment, drawing on areas of relative strength as much as possible’. Recommendations were then produced on the basis of statements with moderate to high agreement, or those with lower agreement where any issues identified by the GC could be easily addressed in the wording of recommendations. The results of round 2, and the resulting recommendations were then presented and discussed. A brief summary of this process is depicted in Table 24 below. A complete list of statements and ratings can be found in Appendix U and blank copies of questionnaires used can be found in Appendix T.
Summary of nominal group technique process followed for the development of recommendations on adaptations to psychological interventions for mental health problems in people with milder learning disabilities.
As above, percentage agreement was calculated for each statement and comments collated. An example of a statement that was rated highly by the committee is: ‘For people with more severe learning disabilities, it may be particularly useful to help to manage the person’s environment to reduce stressors or to help them to manage change’. Recommendations were then produced on the basis of statements with moderate to high agreement, or those with lower agreement where any issues identified by the GC could be easily addressed in the wording of recommendations. The rankings and comments were then presented to the GC members and used to inform a discussion of the issues raised by members’ comments. A second round of ratings was not deemed necessary as it was agreed by the GC that all important issues raised in the GC comments could be addressed in the wording of recommendations. A brief summary of this process is depicted in Table 25. A complete list of statements and ratings can be found in Appendix U and blank copies of questionnaires used can be found in Appendix T.
Summary of nominal group technique process followed for the development of recommendations on adaptations to psychological interventions for mental health problems in people with more severe learning disabilities.
A small number of RCTs (N=10) were found on psychological interventions for the prevention or treatment and management of mental health problems. As many of these studies were pilot studies with very small numbers of participants and the GC were aware of the existence of a number of non-randomised controlled studies, these were also searched for. An existing and recent systematic review which included any controlled studies (randomised or not) was used to identify relevant studies and an update search was conducted (Vereenooghe & Langdon, 2013). All the non-randomised studies identified in this review were on anger or aggression which was addressed in the challenging behaviour guideline (NICE, 2015). The new searches identified 7 additional non-randomised controlled studies which have been included here (n=7).
Most studies were conducted in adults; 1 study was in children on the treatment of PTSD (Holstead & Dalton, 2013). Furthermore, most studies included people with mild to moderate learning disabilities.
For this review, 3 RCTs (N =87) and 1 controlled before-and-after study (N=24) met the eligibility criteria: Matson 1981, Matson & Senatore 1981, Nezu 1991, and Lindsay 2015 (Lindsay et al., 2015; Matson, 1981; Matson & Senatore, 1981; Nezu, 1991). Psychological interventions used in the trials included social skills training, participant modelling/graded exposure, social problem solving followed by assertiveness, assertiveness followed by social problem solving, traditional psychotherapy and CBT. An overview of the trials included can be found in Table 26 and Table 27. Further information about both included and excluded studies can be found in Appendix L.
Study information table for RCTs included in the analysis of psychological interventions for mental health problems.
Study information table for RCTs included in the analysis of psychological interventions for mental health problems.
Summary of findings can be found in Table 29. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Summary of findings table for the analysis of psychological interventions versus control for mental health problems.
Studies included were a combination of prevention and treatment. All participants in these trials had mild to moderate learning disabilities.
Of the 3 RCTs, 2 (Matson & Senatore, 1981; Nezu, 1991) were 3-armed trials. Data from each of the active arms compared with the control were included in the analysis (it was not possible to average the data for both active arms because of the way the data was reported in the studies).
It was not possible to conduct subgroup analyses with the limited number of studies included so a random effects model was used and the outcome was downgraded for inconsistency.
No data were available for the critical outcomes of community participation and quality of life.
| CBT versus control | |
|---|---|
| Total no. of studies (N1) | 1 (24) |
| Study ID | Lindsay 2015 |
| Country | UK |
| Diagnosis/degree of learning disabilities | Mild |
| Age (mean) | 31 |
| Sex (% female) | 50 |
| Ethnicity (% white) | Not reported |
| IQ (mean) | 63.15 |
| Living arrangements | Mixed |
| Coexisting conditions/treatments received | Not reported |
| Targeted behaviour | Mixed2 |
| Treatment length (weeks) | 10.75 (from 8–14) |
| Intervention (mean dose; mg/day) | Weekly trans-diagnostic individual CBT with assistance from significant other person (family member or carer)3 |
| Comparison | Waitlist control (matched to those receiving intervention) |
Note.
Number randomised.
Diagnosis included 8 anxiety, 6 depression, 2 mixed anxiety and depression, 2 interpersonal conflict, 4 bereavement and anxiety and depression, 1 experience of sexual abuse with anxiety and depression.
Length of each weekly session not reported.
All had capacity to consent and participate in the assessment, but no IQ reported.
9 aggressive behaviour, 3 sexually inappropriate behaviour, 3 psychotic/bizarre behaviour, 1 relationship difficulties, 1 self-injury, 1 depression, 1 bulimia and 1 OCD.
There was 1 RCT (N=18) which met the eligibility criteria for this review: Nezu 1991.
An overview of the trial included can be found in Table 30. Further information about both included and excluded studies can be found in Appendix L.
Study information table for RCTs included in the analysis of social problem solving then assertiveness training versus assertiveness training then social problem solving for mental health problems.
Summary of findings can be found in Table 31. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Summary of findings table for the analysis of social problem solving then assertiveness training versus assertiveness training then social problem solving for mental health problems.
No data were available for the critical outcomes of community participation and quality of life.
There was one cohort study (N=28) which met the eligibility criteria for this review: (Beail et al., 2007).
An overview of the study included can be found in Table 32. Further information about both included and excluded studies can be found in Appendix L.
Study information table for cohort studies included in the analysis of psychodynamic psychotherapy at 8 sessions versus 12 sessions versus 24+ sessions for mental health problems.
Summary of findings can be found in Table 33. The full GRADE evidence profiles can be found in Appendices N and O.
Summary of findings table for the analysis of psychodynamic psychotherapy of 8 sessions versus 12 sessions versus 24+ sessions for mental health problems.
The differences are reported narratively only; the study reported that there were no significant differences between global severity scores at all time-points (8, 12 and 24 weeks and at 13 weeks’ follow-up).
No data were available for the critical outcomes of community participation and quality of life.
There was 1, 3-armed RCT (N=84) which met the eligibility criteria for this review: McGillicuddy and Blane (1999). The study compared assertiveness building with modelling and social inference, and a waitlist control group.
An overview of the trial (by pairwise comparison) included can be found in Table 34. Further information about both included and excluded studies can be found in Appendix L.
Study information table for RCTs included in the analysis of psychological interventions versus control for substance misuse.
Summary of findings can be found in Table 35. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Summary of findings table for the analysis of psychological interventions versus control for substance misuse.
The study included participants with and without substance misuse (therefore, was both preventative and treatment) and it was unclear what level of learning disabilities the included participants had.
The arms for the trial were combined in the analysis of psychological interventions with control.
No data were available for the critical outcomes of community participation and quality of life.
There was 1 RCT (N=42) which met the eligibility criteria for this review: McGillicuddy and Blane (1999).
An overview of the trial included can be found in Table 36. Further information about both included and excluded studies can be found in Appendix L.
Study information table for RCTs included in the analysis of assertiveness building versus modelling and social inference for substance misuse.
Summary of findings can be found in Table 37. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Summary of findings table for the analysis of assertiveness training versus modelling and social inference for substance misuse.
No data were available for the critical outcomes of community participation and quality of life.
There were 5 RCTs (N=139) and 2 controlled before-and-after studies (N=118) which were found on the treatment or prevention of anxiety. The included studies were a mixture of treatment, selective prevention, or indicated prevention, covered a mixture of levels of learning disabilities and addressed different type of anxiety (see summary below in Table 38). Only 1 study considered children and young people while all others included adults only.
Summary of included studies on anxiety.
There were 4 RCTs (N=112) and 1 controlled before-and-after study (N=24) which met the eligibility criteria for this review: Hassiotis et al. (2013), Lindsay et al. (1989), Morrison and Lindsay (1997), Peck (1977) and Lindsay et al. (2015).
The studies used a mixture of psychological interventions including CBT, dating skills therapy, and desensitisation. Peck 1977 was a 4-armed trial and Lindsay 1989 was a 3-armed trial.
An overview of the trials included can be found in Table 39 and Table 40. Further information about both included and excluded studies can be found in Appendix L. Summary of findings can be found in Table 41. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Study information table for RCTs included in the analysis of psychological interventions for anxiety symptoms in people with mild to moderate learning disabilities.
Study information table for controlled before-and-after studies included in the analysis of psychological interventions for anxiety symptoms in people with mild to moderate learning disabilities.
Summary of findings table for the analysis of psychological interventions for anxiety symptoms in people with mild to moderate learning disabilities.
It was not possible to present the results for Peck 1976 (due to lack of variance reported in the studies) so data is presented here narratively: contact desensitisation appeared better than placebo after 5 weeks of treatment on the Behaviour Avoidance Test, Behaviour Checklist ratings, Fear Thermometer, and the Modified Fear Survey. Neither vicarious symbolic desensitisation nor systematic desensitisation appeared better than placebo.
Considerable heterogeneity was found in the outcome on anxiety in the RCTs but there were too few studies to assess the reasons for this. As such, a random-effects model was used and the outcome was downgraded for inconsistency.
No data were available for the critical outcome of quality of life.
Hassiotis et al. (2013) conducted face-to-face or telephone interviews with 13 patients in the CBT arm at least 1 month after completion. The questionnaire included 6 open-ended questions. The study also extracted qualitative data on the patient experience from discussions with professionals from community teams where participants were recruited.
There were 7 themes presented from the interviews, and they are summarised in Table 42.
Summary of themes from experience of care with CBT.
The study is of high quality overall (83.3% of criteria met), with mixed reporting of aspects of rationale and the methods used but well-reported analysis and ethics. This study appears to show that manualised individual CBT is generally valued as a useful intervention for symptoms of anxiety and depression and is both appropriate for use with and well-tolerated by service users with learning disabilities.
There were 2 RCTs (N = 60) which met the eligibility criteria for this review: Lindsay et al. (1989) and Morrison and Lindsay (1997).
An overview of the trials included can be found in Table 43 Further information about both included and excluded studies can be found in Appendix L.
Study information table for trials included in the analysis of relaxation training versus control for anxiety symptoms in people with moderate to severe learning disabilities.
Summary of findings can be found in Table 44. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Summary of findings table for the analysis of relaxation training versus control for anxiety symptoms in people with moderate to severe learning disabilities.
The RCT by Lindsay 1989 had 4 active arms with varying types of relaxation therapy, and these are presented separately in the analysis of relaxation therapy versus control.
There was inconsistency in the results for the anxiety symptom outcome for the subgroup of individual group training and was not possible to explore in sensitivity or subgroup analyses due to the small number of trials. As a result, a random-effects model was used and the outcome was downgraded for inconsistency.
No data were available for the critical outcomes of community participation and quality of life.
There was 1 RCT (N=27) which met the eligibility criteria for this review: Valenti-Hein et al. (1994).
An overview of the trial included can be found in Table 45. Further information about both included and excluded studies can be found in Appendix L.
Study information table for RCTs included in the analysis of dating skills versus control for social anxiety symptoms.
Summary of findings can be found in Table 46. The full GRADE evidence profiles and associated forest plots can be found in Appendix N and Appendix O.
Summary of findings table for the analysis of dating skills versus control for social anxiety symptoms.
No data were available for the critical outcomes of community participation and quality of life.
There was 1 controlled before-and-after study (N=88) which met the eligibility criteria for this review: Holstead and Dalton (2013).
An overview of the trial included can be found in Table 47. Further information about both included and excluded studies can be found in Appendix L.
Study information table for controlled before-and-after studies included in the analysis of CBT versus applied behaviour analysis for PTSD.
Summary of findings can be found in Table 48. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Summary of findings table for the analysis of CBT versus applied behaviour analysis for PTSD.
No data were available for the critical outcomes of community participation and quality of life.
There were 3 RCTs (N=130) and 3 controlled before-and-after studies (N=130) which met the eligibility criteria for this review: McCabe et al. (2006), McGillivray et al. (2008), Hassiotis et al. (2013), Hartley et al. (2015), Lindsay et al. (2015), and McGillivray and Kershaw (2013).
An overview of the trials included can be found in Table 49 and Table 50. Further information about both included and excluded studies can be found in Appendix L.
Study information table for RCTs included in the analysis of cognitive behavioural training versus control for depressive symptoms.
Study information table for controlled before-and-after studies included in the analysis of cognitive behavioural training versus control for depressive symptoms.
Summary of findings can be found in Table 51. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Summary of findings table for the analysis of CBT versus control for depressive symptoms.
All but 1 study were conducted in participants with depressive symptoms or diagnosed with depression so would be considered to be treatment studies. A different study included participants who were at risk for depression (McGillivray & Kershaw, 2013); this was defined as the presence of at least 4 items/symptoms on the Depression Screening Checklist (with 22 items/symptoms/risk factors) and at least 1 mood and/or activities item (Beck Depression Inventory, 1996 revision [BDI-II]: 34% were minimally, 40% mildly, 19% moderately and 7% severely depressed).
Most studies included both patients with mild and moderate learning disabilities, though 1 only included mild learning disabilities. Those that included both mild and moderate learning disabilities did not present the results for mild and moderate learning disabilities separately so it was not possible to see if the results differed by degree of learning disabilities.
While most studies reported outcomes immediately after treatment, 2 studies reported outcomes after a period of follow-up and this was used in the analyses (Hassiotis 2013, Hartley 2015). A sensitivity analysis using the outcomes immediately after treatment (and more comparable with the other studies included in the analysis) had similar results.
No data were available for the critical outcome of quality of life.
Hassiotis 2013 conducted qualitative interviews of the patient experience of CBT. These are presented above in the section on anxiety (see section 1.1.1.1.1).
There was 1 controlled before-and-after study (N=70) which met the eligibility criteria for this review: McGillivray (2015).
An overview of the trial included can be found in Table 52. Further information about both included and excluded studies can be found in Appendix L.
Study information table for controlled before-and-after studies included in the analysis of CBT versus behavioural or cognitive strategies only for depressive symptoms.
Summary of findings can be found in Table 53 and Table 54. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Summary of findings table for the analysis of CBT versus behavioural strategies only for depressive symptoms.
Summary of findings table for the analysis of CBT versus cognitive strategies only for depressive symptoms.
The study included patients with mild learning disabilities only.
No data were available for the critical outcomes of community participation and quality of life.
There was 1 cohort study (N=18) which met the eligibility criteria for this review: (Beail, 2001).
An overview of the trial included can be found in Table 55. Further information about both included and excluded studies can be found in Appendix L.
Study information table for cohort studies included in the analysis of psychodynamic psychotherapy versus control for sexually inappropriate behaviour.
Summary of findings can be found in Table 56. The full GRADE evidence profiles and associated forest plots can be found in Appendices N and O.
Summary of findings table for the analysis of psychodynamic psychotherapy versus control for sexually inappropriate behaviour.
The GC endorsed statements supporting the need to:
The GC endorsed statements stating that:
The GC endorsed statements stating that it was important to:
The GC endorsed statements stating that it may be helpful to:
The GC endorsed statements stating that it was important or helpful to:
The systematic search of the literature identified 1 study that assessed the cost effectiveness of psychological interventions for the management of mental health problems in adults with learning disabilities, which was conducted in the UK (Hassiotis et al., 2013). Details on the methods used for the systematic review of the economic literature are described in Chapter 3; full references and evidence tables for all economic evaluations included in the systematic literature review are provided in Appendix R. Completed methodology checklists of the studies are provided in Appendix Q. Economic evidence profiles of studies considered during guideline development (that is, studies that fully or partly met the applicability and quality criteria) are presented in Appendix S.
Hassiotis and colleagues (2013) evaluated the cost effectiveness of manualised individual CBT added to treatment as usual versus treatment as usual alone, for the management of adults with mild to moderate learning disabilities who had a mood disorder or symptoms of depression and/or anxiety. The economic analysis was conducted alongside a feasibility RCT (Hassiotis 2013, N=32). CBT consisted of 16 weekly 1-hour sessions. Treatment as usual comprised follow-up appointments of service users with their clinicians or care coordinators as described in their care plans. The perspective of the analysis was the NHS and social care services. Costs consisted of intervention costs (CBT), inpatient and outpatient care, emergency visits, community care, day care and paid care. National unit costs were used. The primary measures of outcome were the mean change in the Beck Depression Inventory-Youth (BDI-Y) and the Beck Anxiety Inventory-Youth score from baseline to endpoint. The duration of the study was 16 weeks.
Before treatment, the total mean cost per person was £4,551 (SD £7,568) for the CBT group and £2,420 (SD £6,289) for the control group, resulting in a cost difference of £2,131 (2010 prices); over the 16 weeks of treatment the total mean cost per person was £7,327 (SD £8,007) for the CBT group and £1,677 (SD £2,415) for the control group, so that the cost difference between the 2 arms of the trial was £5,650. The paper did not report whether the cost differences were statistically significant, but it is very unlikely they were due to the small number of participants and the wide confidence intervals around the mean costs. In terms of outcomes, changes in the BDI-Y score favoured CBT, while changes in the Beck Anxiety Inventory-Youth score favoured treatment as usual; none of the changes were statistically significant. Based on these results, it is unclear whether the addition of CBT on treatment as usual is a cost-effective option for the management of mood disorders or symptoms of depression and/or anxiety in people with mild to moderate learning disabilities.
The study is partially applicable to the NICE decision-making context, as no quality adjusted life years (QALYs), which is the preferred outcome measure by NICE for economic analyses, were estimated and therefore it is difficult to make judgements on the cost effectiveness of the intervention using the NICE cost effectiveness threshold. Moreover, the study population, which consisted of people with mild to moderate learning disabilities who had a mood disorder or symptoms of depression and/or anxiety, comprises only a sub-group within the population of people with learning disabilities and mental health problems. Therefore, the results of this study may not be applicable to people with more severe learning disabilities, or people with learning disabilities experiencing different types of mental health problems. The economic study is characterised by very serious limitations, as it was conducted alongside a feasibility RCT, and therefore had a very small study sample (N=32), a short time horizon of 16 weeks, and, unsurprisingly, high uncertainty around costs and outcomes. For this reason, it was not considered further at formulation of recommendations.
| Recommendations | Delivering psychological interventions for mental health problems in people with learning disabilities
|
|---|---|
| Relative values of different outcomes | The GC discussed the importance and relevance of various outcomes in the evidence when assessing the effectiveness of interventions at preventing or treating mental health problems. In addition to the effect on the mental health problem which was the aim of the intervention (for dementia, this was cognitive and adaptive function as well as psychopathology), the GC were of the view that quality of life, and community participation and meaningful occupation were particularly critical outcomes which they wished to consider in the literature. The GC noted in particular the difficulties with measuring self-reported outcomes in this population, given communication needs and cognitive impairments so there should be caution in the interpretation of these outcomes from the trials. Reported outcomes from multiple sources may be helpful (for example, teachers, parents) in addressing this issue. Additional important outcomes included problem behaviours, adaptive functioning such as communication skills, service user or carer satisfaction or experience of care, carer health and quality of life, adverse effects of interventions, rates of placement breakdown (including out-of-area placements or rates of restrictive interventions), psychiatric hospital admissions (including length of stay or other outcomes related to admission), as well as offending or re-offending. |
| Trade-off between clinical benefits and harms | The group noted that while there was more evidence on psychological interventions than other types of interventions, there is still relatively little evidence of psychological interventions in people with learning disabilities. In addition, all but 1 study (Holstead & Dalton, 2013) was in adults so there is a particular lack of evidence on children and young people. They also noted that most studies included people with mild to moderate learning disabilities, only 1 study (Lindsay et al., 1989) on relaxation therapy for anxiety included some people with severe learning disabilities. They noted the overall lack of evidence on community participation and meaningful occupation and quality of life in this area. They also noted that there was very little evidence on harms from psychological treatment but that there was unlikely to be major harms from most of the types of treatments considered in this section. There is the possibility of harm from desensitisation or exposure therapy but there was no evidence of this. Mixed mental health problems The group noted that psychological interventions, in general, appear to be better than no treatment or waitlist control at reducing mental health problems in people with mild to moderate learning disabilities but this is based on a small number of studies. One RCT compared the difference between social problem solving then assertiveness training with assertiveness training then social problem solving and a non-randomised study compared different lengths of psychodynamic psychotherapy but there was no difference between groups. Substance misuse There was 1 study which was a 3-armed trial on prevention or treatment of substance misuse, and it considered assertiveness training with modelling plus social inference or control. However, it showed very little difference between arms and the sample was small so may have not been able to detect differences. Anxiety disorders The evidence on psychological interventions for anxiety symptoms in mild- to-moderate learning disabilities was inconclusive. The RCTs showed that there may be an effect of psychological interventions over control in terms of anxiety symptoms; the controlled before-and-after study was more uncertain. However, see notes below on quality considerations. There was some RCT evidence that there were less people in paid employment with psychological interventions compared with control but more were in voluntary work. However, there was much uncertainty around these estimates so it was difficult to draw any conclusions. There was some evidence to suggest that contact desensitisation is better than placebo at reducing fears but this was based a small and older study. The evidence on psychological interventions for moderate to severe learning disabilities was only on relaxation therapy and appeared to be consistent in showing that relaxation is better than placebo at reducing anxiety symptoms. While the evidence on relaxation was from small studies and was of limited quality, the results were considered to be clinically important and precise. As such, the GC agreed that they could make a weak recommendation on relaxation therapy; while the evidence supported the use of this in moderate and severe learning disabilities, the GC considered that it is reasonable to expect that it would be similarly effective in mild learning disabilities so did not limit the population in the recommendation. There was 1 trial on social anxiety which found no difference between a dating skills programme and a control in a group of people with mild to moderate learning disabilities. The trial appears to have included people with borderline intellectual functioning but it’s not clear how many would be considered to have learning disabilities (and, thus, not be the population of interest) as the study used an atypical measurement of intellectual functioning on a 3-point scale. The GC agreed that more research is needed on more targeted and structured interventions for anxiety and that these should involve a social component. They considered that these types of interventions are likely to be quite useful in people with learning disabilities but there was no data to support this. Post-traumatic stress disorder There was 1 controlled before-and-after trial which compared CBT versus ABA for children with trauma which appeared to show that ABA is favourable over CBT on most subscales on the Achenbach assessment tool. However, see quality considerations below. As such, the GC did not consider that they could draw any conclusions from this evidence on the treatment of PTSD. They were particularly concerned about the lack of evidence on PTSD in people with learning disabilities as they are a vulnerable group who are at risk of trauma, abuse and neglect. Depressive symptoms All the evidence on psychological interventions for depression was for CBT (adapted for learning disabilities) in people with mild to moderate learning disabilities. While the trials included appeared to have some variations in the adaptations to the intervention and suffered from the same issues as most other studies in learning disabilities (being mostly pilot feasibility studies so small in size and leading to imprecision in estimates), the evidence appeared to suggest that CBT may be beneficial for depression. There was some RCT evidence (from the same study reported above for anxiety) that there were less people in paid employment with CBT compared with control but more were in voluntary work. However, there was much uncertainty around these estimates so it was difficult to draw any conclusions. The effects on social skills were less clear but the GC considered that this may be due to the variations in CBT across the studies (see further details below under quality). Given the GC’s expert knowledge and some demonstrated evidence of a benefit for CBT over control, the GC decided to make a recommendation on adapted CBT for depression. However, as the evidence was from small studies and very low quality (see ‘Quality of the evidence’ overleaf), they were not confident enough in the results to make a strong recommendation so recommended that CBT be considered for use in depression. There was little difference between CBT and behavioural strategies only or CBT compared with cognitive strategies only but the evidence was based on one small controlled before-and-after study. Sexually inappropriate behaviour One very small observational study which considered psychodynamic psychotherapy against no treatment in a population of offenders (13 of 18) reported reduced recidivism in the treatment group but there was a lot of uncertainty in the results, possibly due to the small numbers of patients in the study, so it was not possible to draw any conclusions. No evidence was found on the use of psychological interventions for the treatment and prevention of other mental health problems in people with learning disabilities, such as substance misuse (other than alcohol misuse), eating disorders, personality disorders, or serious mental illness such as psychosis or schizophrenia. |
| Trade-off between net health benefits and resource use | Limited evidence on the cost effectiveness of psychological interventions for adults was inconclusive and characterised by very serious limitations and therefore was not considered further. The GC considered the economic consequences arising from the presence of mental health problems in people with learning disabilities that is associated with consumption of extra healthcare resources. The GC also considered the impact of mental health problems on the (HRQoL) of people with learning disabilities, their family and carers and concluded that provision of effective psychological interventions for the prevention and management of mental health problems is likely to improve the HRQoL of service users and carers and reduce healthcare costs resulting from the management of mental health problems in more resource-intensive settings, such as secondary care. The GC took into account the fact that provision of psychological interventions to people with learning disabilities may be more resource- intensive compared with provision of psychological interventions in populations without learning disabilities, and this may have implications for the cost effectiveness of such interventions. Nevertheless, the GC agreed that adaptations of psychological interventions were necessary for this population in order to achieve a positive outcome. The GC considered also issues relating to equality, and agreed that psychological interventions for the prevention and/or management of mental health problems that have been shown to be cost-effective in populations without learning disabilities should also be offered to people with learning disabilities, following necessary adaptations. |
| Quality of evidence | The overall quality of the evidence was low to very low quality. The GC were particularly concerned about the nature of much of the existing RCT data, particularly as most are feasibility studies and are not likely to be powered appropriately to determine effectiveness against a comparator. However, they also appreciated the additional risk of bias by considering controlled before and after studies. For the most part, the controlled before and after studies were consistent with the results from the RCTs. There was inconsistency within the analyses for a number of comparisons:
The GC noted again that many of outcome measurements used in the included trials were experimental or bespoke measures or had not been validated in this population (such as the BDI) and that there should be caution in the interpretation of some of the outcomes from the trials. As a result of the quality of the evidence, the GC did not have confidence in the results from the evidence. |
| Other considerations | As a result of limited quality evidence the group agreed to develop a set of general principles for adapting psychological treatments for people with learning disabilities. The GC decided on the basis of the outcome of the nominal group technique to develop recommendations in the following areas:
While the evidence for CBT for depression was a bit clearer, the evidence for psychological interventions for anxiety disorders (including PTSD) was less clear. As such, the group made a recommendation for future research into interventions for anxiety. The group also noted the lack of evidence on psychological interventions for conditions other than common mental health problems such as eating disorders, schizophrenia, and bipolar disorder so recommended more research in these areas, as well. The group noted that the existing evidence on psychological interventions was in adults but decided, in the absence of evidence for children, that the recommendations should also apply to children to prevent children from falling between the gaps and ensure they receive treatment, when needed. However, the group recommended that future research on individual treatments for children should be conducted in order to inform specific recommendations on treatment for children. |
| Recommendations | Interventions for mental health problems in people with learning disabilities
|
|---|---|
| Relative values of different outcomes | The GC discussed the importance and relevance of various outcomes in the evidence when assessing the effectiveness of interventions at preventing or treating mental health problems. In addition to the effect on the mental health problem which was the aim of the intervention (for dementia, this was cognitive function as well as psychopathology), the GC were of the view that quality of life, and community participation and meaningful occupation were particularly critical outcomes which they wished to consider in the literature. The GC noted in particular the difficulties with measuring self-reported outcomes in this population, given communication needs and cognitive impairments so there should be caution in the interpretation of these outcomes from the trials. Reported outcomes from multiple sources may be helpful (for example, teachers, parents) in addressing this issue. Additional important outcomes included problem behaviours, adaptive functioning such as communication skills, service user or carer satisfaction or experience of care, carer health and quality of life, adverse effects of interventions, rates of placement breakdown (including out-of-area placements or rates of restrictive interventions), psychiatric hospital admissions (including length of stay or other outcomes related to admission), as well as offending or re-offending. |
| Trade-off between clinical benefits and harms | Please see this section for each intervention section (5.2.7.1, 5.3.5, 6.3, 7.5, 7.7.4, and 7.8.5) |
| Trade-off between net health benefits and resource use | Please see this section for each intervention section (5.2.7.1, 5.3.5, 6.3, 7.5, 7.7.4, and 7.8.5) |
| Quality of evidence | Please see this section for each intervention section (5.2.7.1, 5.3.5, 6.3, 7.5, 7.7.4, and 7.8.5) |
| Other considerations | With the exception of some specific psychological interventions for which the GC were able to make recommendations from the available evidence, there was a paucity of high quality research to inform recommendations on interventions for people with learning disabilities and mental health problems. The GC agreed that people with learning disabilities should be offered the same services and interventions as people without learning disabilities and judged that it was appropriate to refer into other NICE guidelines on mental health problems. However, they appreciated that there are some additional considerations that need to be made when considering interventions for people with learning disabilities. Some of these considerations are covered in recommendations developed through formal consensus on adapting psychological interventions as well as on the use of pharmacological interventions. The GC developed a general set of principles to inform the approach to assess the relevance of other mental health guidelines for the treatment of people with learning disabilities and mental health problems. For people with milder learning disabilities, the GC considered that all NICE guideline recommendations were relevant. For moderate and severe learning disabilities, the GC took the view that the general principles they developed were the best guide to the appropriate use and adaptation of existing guidance. The GC noted that variations exist within the population of people with learning disabilities (in the nature of learning disabilities, the presentation of mental health problems, and different coexisting conditions which occur in many people with learning disabilities) and this means that individual clinicians will have to make a judgment on the relevance of a particular recommendation in NICE guidelines on mental health problems, depending on the specific needs of the individual. They also agreed that special considerations about the ideal treatment setting for a person should be made on an individual basis. Specific recommendations around communication and decision-making (with respect to consent and capacity) are important considerations which have been highlighted elsewhere in this guideline (see 8.2.6.1 and 8.2.6.2). The GC agreed that patient-reported outcome and experience measures are an important evaluative component of the efficacy and acceptability of any intervention provided to prevent or treat mental health problems. There is currently a dearth of such instruments, and therefore the GC agreed to make a research recommendation for the development of these tools. |
Note.
GC = Guideline Committee; NICE = National Institute for Health and Care Excellence.
For children and young people with learning disabilities, what psychological interventions (such as cognitive behavioural therapy and interpersonal therapy) are clinically and cost effective for treating internalising disorders?
For adults with milder learning disabilities, what is the clinical and cost effectiveness of psychological interventions such as cognitive behavioural therapy (modified for people with learning disabilities) for treating depression and anxiety disorders?
For people with more severe learning disabilities, what is the clinical and cost effectiveness of psychosocial interventions to treat mental health problems?
What experience do people with learning disabilities have of services designed to prevent and treat mental health problems and how does this relate to clinical outcomes?
Develop patient-reported outcome and experience measures for use with people with learning disabilities and mental health problems.
The review protocol summary, including the review question and the eligibility criteria used for this section of the guideline, can be found in Table 57. A complete list of review questions and review protocols can be found in Appendix F; further information about the search strategy can be found in Appendix H.
Clinical review protocol summary for the review on the prevention of mental health problems.
A number of RCTs were found on the use of parent training interventions; however there were no studies found on staff training or training for other family carers which were focused on mental health problems in people with learning disabilities.
The review completed for the guideline on behaviour challenges on parent training was adapted for use in this guideline (NICE, 2015). However, some additional information was extracted from the papers such as the underlying cause of learning disabilities, any coexisting conditions/treatments received, whether or not the intervention was tailored for learning disabilities (see the study information tables below), and additional outcome measures to indicate mental health (when reported). There was 1 analysis from the review for parent training versus any control that was amended to include these different outcome measures that the GC considered as measures of mental health, rather than just behaviour that challenges (that is, total scores on the Developmental Behaviour Checklist [DBC] rather than the disruptive behaviour subscale on this tool). The guideline on behaviour that challenges also included pairwise comparisons of studies comparing different types of parent training (individual versus group, parent training plus optimism training versus parent training alone, and enhanced versus standard parent training) but these have not been considered here as the studies included did not report an outcome that would be considered a measure of mental health as a primary outcome.
The GC did not consider it necessary to go down the evidence hierarchy for parent training as the existing RCT evidence was considered to be adequate to support recommendations. Furthermore, the group did not update the challenging behaviour review as it was considered unlikely that new research would alter the findings significantly.
There were 15 RCTs (N=819) that met the eligibility criteria for this review: Aman (2009), Bagner and Eyberg (2007), Brightman et al. (1982), Hand et al. (2012), Leung et al. (2013), McIntyre (2008), Oliva et al. (2012), Plant and Sanders (2007), Prieto-Bayard and Baker (1986), Reitzel et al. (2013), Roberts et al. (2006), Roux et al. (2013), Sofronoff et al. (2011), Tellegen and Sanders (2013) and Whittingham et al. (2009). Of the eligible studies, 13 included sufficient data to be included in a meta-analysis, 1 trial (Prieto-Bayard 1986) included no critical outcome data and was therefore excluded (N=20) and 1 trial (Brightman 1982; N=66) included critical outcomes that could not be included in the meta-analysis because of the way the data had been reported. A brief narrative synthesis of Brightman (1982) is given to assess whether the findings support or refute the meta-analysis. An overview of the 14 trials included in the meta-analysis can be found in Table 58.
Study information table for trials included in the analysis of parent training versus any control.
Further information about both included and excluded studies can be found in the challenging behaviour guideline (NICE, 2015).
Summary of findings can be found in Table 59. The full GRADE evidence profiles and any amended forest plots can be found in Appendix N and Appendix O, respectively.
Summary of findings table for the analysis of parent training versus control.
While some studies did not report the level of learning disabilities of the included participants, others included patients across differing levels of disabilities without reporting separate results by level of disabilities. As a result, it was not possible to group the results by level of disabilities.
There were 3 studies which considered mixed populations of learning disabled and non-learning disabled participants (Aman 2009; Tellegen 2014; Whittingham 2009). To explore the robustness of the findings, a second sensitivity analysis excluding these 3 studies was conducted. All but 1 effect remained consistent with the main analysis (the removal of Aman 2009 led to insufficient evidence to assess adaptive functioning).
Subgroup analysis was carried out to compare the effectiveness of parent training delivered to individuals with that of parent training delivered to groups. Both subgroups were shown to be equally effective at reducing mental health/behavioural symptoms.
Subgroup analyses were also conducted to see if studies explicitly addressing parents of children with specific symptoms (treatment) produced different results than those directed at any parent of a child with learning disabilities, whether or not they had any existing symptoms or diagnoses (mixed prevention or treatment). There was no significant difference in the effect in the subgroups. Subgroup analyses comparing studies with programs explicitly tailored for parents of children with learning disabilities with those that were not tailored also showed no difference between groups. All subgroup analyses showed parent training as effective at improving mental health/behavioural outcomes for all subgroups.
No data were available for the critical outcomes of quality of life or community participation and meaningful occupation.
No economic evidence was identified for family carer or staff training aimed at preventing and/or managing mental health problems in people with learning disabilities. Details on the methods used for the systematic review of the economic literature are described in Chapter 3. Because of lack of direct economic evidence, the GC considered economic evidence that was reported in the NICE Guideline on Challenging Behaviour and Learning Disabilities (NICE, 2015) for children and young people with learning disabilities and behaviour that challenges. An economic model was developed for that guideline, which assessed the cost effectiveness of parent training for the management of behaviour that challenges in children and young people with learning disabilities. A summary of the methods and the results of this model are presented in this section, as the GC considered the findings of this economic analysis as indirect evidence on the cost effectiveness of parent training for the management of mental health problems in children and young people with learning disabilities. The completed methodology checklist, the evidence table and the economic profile of this economic study are provided in Appendix Q, Appendix R and Appendix S, respectively.
The economic analysis developed to inform recommendations in the NICE clinical guideline on Challenging Behaviour and Learning Disabilities (NICE, 2015) compared parent training with waitlist for the management of behaviour that challenges in children and young people with learning disabilities. The analysis considered group parent training because available evidence suggested that there was no difference in the clinical effectiveness between individual and group parent training; therefore group parent training was selected for modelling as it is more cost-effective than parent training delivered individually (because the intervention cost is lower). Waitlist was selected as the comparator as this was the most common control used in the RCTs that informed that economic analysis. In those RCTs that did not use waitlist as the comparator, parent training was predominantly provided in addition to treatment as usual versus treatment as usual alone, so the control intervention did not incur any extra costs.
The economic model, which had the form of a decision-tree, followed hypothetical cohorts of families of children and young people with learning disabilities and behaviour that challenges, who either received group parent training for 9 weeks or were included in a waitlist. Families of children and young people whose symptoms improved at the end of the 9 weeks received 2 booster sessions; children and young people with improved symptoms could relapse over the following year. Children and young people whose behaviour did not improve at the end of the 9 weeks were conservatively assumed to retain behaviour that challenges over the following year. The time horizon of the model was 61 weeks (9 weeks of treatment and 52 weeks of follow-up). The analysis adopted the perspective of the NHS and personal social services. Costs consisted of intervention costs only, as no data on costs associated with behaviour that challenges in children and young people with learning disabilities were identified in the relevant literature. The measure of outcome was the QALY.
Efficacy data regarding the relative effect of parent training versus waitlist and the baseline effect of waitlist were taken from 8 RCTs on parent training that were included in the guideline systematic review, which reported outcomes in the form of improvement in behaviour that challenges regarding its severity. Improvement was defined as a clinically significant change score in 1 of the following scales: the Eyberg Child Behavior Inventory (ECBI) – Problem, the Child Behavior Checklist – Externalising behaviour, or the Developmental Behavior Checklist – Total Behavior Problem (DBC-TBPS). The probability of relapse was based on the GC expert opinion, due to lack of relevant data in the literature. Utility data used for the estimation of QALYs were taken from Tilford et al. (2012), following a systematic review of relevant literature. The study reported utility data for children with autism in the US, derived from their parents’ responses to the Health Utility Index Mark 3. The economic analysis used utility scores reported in that study for different levels of hyperactivity as a proxy for changes in behaviour that challenges in children and young people with learning disabilities. The analysis conservatively assumed that at initiation of treatment the (HRQoL) of children and young people corresponded to moderate levels of hyperactivity that improved to mild symptoms following response to treatment
The intervention cost of parent training was calculated by combining relevant resource use (based on data reported in the 8 RCTs included in the guideline systematic review that were considered in the economic analysis) with respective national unit costs (Curtis, 2013). The economic analysis modelled parent training comprising 8 group sessions lasting 2 hours each plus 2 booster group sessions of the same duration provided to families whose children showed improvement in their behaviour; each group was formed by 10 families and was run by a clinical psychologist Band 8a and a mental health nurse Band 5 (according to Agenda for Change of the July 2012-June 2013 NHS staff earnings estimates for qualified Allied Health Professionals and qualified nurses, respectively), who acted as co-facilitator. The intervention cost of waitlist was 0.
According to the results, provision of parent training resulted in 1.33 additional QALYs per 100 children and young people with learning disabilities and behaviour that challenges, compared with waitlist, at an additional cost of £36,219 (2013 prices). The incremental cost-effectiveness ratio (ICER) of parent training versus waitlist was £27,148/QALY, which is above the lower (£20,000/QALY) but below the upper (£30,000/QALY) NICE cost-effectiveness threshold. The probability of parent training being cost-effective relative to waitlist under the NICE lower and upper cost effectiveness thresholds was 0.29 and 0.52, respectively. When a lower risk of relapse over 1 year was assumed for parent training (that is, 0.40 instead of 0.50), its ICER versus waitlist fell at £24,895/QALY and its probability of being cost-effective under the lower and upper NICE cost effectiveness thresholds rose at 0.34 and 0.56, respectively. When the HRQoL of children and young people was assumed to correspond to severe hyperactivity at initiation of treatment, the ICER versus waitlist became £13,037/QALY; the probability of parent training being cost-effective under the lower and upper NICE cost effectiveness thresholds was 0.81 and 0.93, respectively, under this scenario.
The results of this analysis indicated that parent training might be marginally cost-effective for the management of behaviour that challenges in children and young people with learning disabilities, although the cost effectiveness of parent training improved when the long-term benefit was assumed to be better retained, and, in particular, when the severity of symptoms was higher at initiation of treatment, as there was more scope for improvement in terms of the children’s and young people’s HRQoL.
The analysis is only partially applicable in the context of this guideline, as the study population of the economic model was children and young people with learning disabilities and behaviour that challenges. The analysis is characterised by potentially serious limitations, including lack of follow-up data (beyond 9 weeks) and omission of costs associated with the presence of behaviour that challenges in children and young people with learning disabilities due to lack of any relevant data. Moreover, the analysis did not consider other benefits to the families and carers associated with group parent training, arising from meeting with other families and carers with similar experiences, sharing ideas and receiving peer support. It should also be noted that the economic analysis modelled only group parent training; individual parent training was expected to be less cost-effective, as it was no more effective and incurred higher intervention costs. However, there may be instances where group parent training is not available or not appropriate for some sub-populations, and individual parent training may be the only treatment option to offer.
The GC of this guideline considered this evidence and decided that it is adequately applicable to the population and context of this guideline.
| Recommendations |
|
|---|---|
| Relative values of different outcomes | The GC discussed the importance and relevance of various outcomes in the evidence when assessing the effectiveness of interventions at preventing or treating mental health problems. In addition to the effect on the mental health problem which was the aim of the intervention (for dementia, this was cognitive function as well as psychopathology), the GC were of the view that quality of life, and community participation and meaningful occupation were particularly critical outcomes which they agreed to consider in the literature. The GC noted in particular the difficulties with measuring self-reported outcomes in this population, given communication needs and cognitive impairments so there should be caution in the interpretation of these outcomes from the trials. Reported outcomes from multiple sources may be helpful (for example, teachers, parents) in addressing this issue. Additional important outcomes included problem behaviours, adaptive functioning such as communication skills, service user or carer satisfaction or experience of care, carer health and quality of life, adverse effects of interventions, rates of placement breakdown (including out-of-area placements or rates of restrictive interventions), psychiatric hospital admissions (including length of stay or other outcomes related to admission), as well as offending or re-offending. |
| Trade-off between clinical benefits and harms | Parent training appeared consistently to improve problem behaviour outcomes for children with learning disabilities with mixed degrees over learning disabilities over control but there was a lack of long-term data and data on harms with parent training. The GC recognised the potential value of early interventions because they equip parents to better manage behaviour so that they may not develop into long-term problems resulting in greater burden for the person, the family and the wider service system. In doing so the GC drew on their experience that parent training is common practice for children with behavioural problems and other neurodevelopmental disorders (for example, ADHD and autism). In particular, this knowledge was used to provide advice about the number of sessions and other aspects of parent- training programmes. The GC noted that parent training also has a positive effect on parental mental health and well-being. These outcomes were not extracted for the challenging behaviour guideline which was adapted for use here, but were extracted as part of the review conducted on interventions for family and carers. The participants in the studies included a range of behavioural problem and some had a diagnosis of oppositional defiance disorder but in most cases the studies were assessed as being selective or indicated prevention studies. As such, the GC agreed that they should recommend the use of parent training programmes for both prevention and treatment. |
| Trade-off between net health benefits and resource use | Evidence suggests that group parent training is potentially cost-effective for the management of problem behaviour in children and young people with learning disabilities, especially in children and young people with more severe levels of behaviour that challenges at initiation of treatment. The GC considered this evidence and decided that it was applicable to the population of children and young people with learning disabilities and mental health problems. In addition, the GC noted the limitations of this evidence, such as the lack of long-term clinical data and the consideration of intervention costs only. The economic evidence was based on an economic model that had considered group parent training, because available evidence suggested that there was no difference in the clinical effectiveness between individual and group parent training; therefore group parent training was selected for modelling as it would be more cost-effective than parent training delivered individually (because the intervention cost is lower). The GC noted the evidence suggesting no difference in clinical effectiveness between individual and group mode of delivery, in combination with the lower intervention costs associated with provision of parent training delivered in groups, and decided to recommend group parent training. The GC considered other benefits resulting from group psychological interventions, such as meeting with other parents and carers experiencing similar situations and exchanging such experiences, sharing ideas and receiving peer support, which was not captured within the existing evidence. The economic evidence was based on provision of 8 sessions of group parent training plus 2 booster sessions following symptom improvement. In total, this required 20 hours of staff’s direct time. This estimate of resource use resulted from summarising the reported resource use in the trials that provided the clinical data that informed the economic model. The GC noted that the NICE guideline on people with learning disabilities and behaviour that challenges (NICE, 2015)recommended group parent training typically consisting of 8 to 12 sessions lasting 90 minutes each. This translates into 12–18 hours of staff’s time, which does not exceed the resource use considered in the economic model, and therefore ensures that provision of parent training is cost-effective. The GC considered the number and duration of parent training sessions recommended in the NICE guideline on people with learning disabilities and behaviour that challenges and found it reasonable from a clinical and cost-effectiveness perspective. Moreover, the GC considered that a percentage of children with learning disabilities have both behaviour that challenges and mental health problems and parent training might be used to address both. Therefore, the GC decided to recommend the same number and duration of group parent training sessions for children with learning disabilities and mental health problems, given also that the same economic evidence base was used in both areas (i.e. children with learning disabilities and behaviour that challenges and children with learning disabilities and mental health problems). |
| Quality of evidence | Unlike some of the other evidence on psychological interventions, the evidence on parent training was moderate to very low quality. The studies on parent training compared with a control were larger than many of the other studies and, partly as a consequence of this, the outcomes did not suffer from imprecision. As the quality of the evidence for parent training was better than in most other areas in this guideline, the GC had relative confidence in the results from the evidence. |
| Other considerations | The GC agreed that getting feedback from parents is good practice so they recommended this. This can be informative for any issues like why people may have dropped out. |
All rights reserved. NICE copyright material can be downloaded for private research and study, and may be reproduced for educational and not-for-profit purposes. No reproduction by or for commercial organisations, or for commercial purposes, is allowed without the written permission of NICE.
Your browsing activity is empty.
Activity recording is turned off.
See more...
