Can cognitive remediation improve subsequent response to low-intensity cognitive behaviour therapy for psychosis in people with schizophrenia?

Evidence suggests that both cognitive behavioural therapy for psychosis (CBTp) and cognitive remediation (CR) are helpful for persons with schizophrenia. The most recent clinical practice guidelines for the management of schizophrenia and related disorders by the Royal Australian and New Zealand College of Psychiatrists recommend CR and CBTp (Galletly et al., 2016). Although both therapies can potentially improve functioning in persons with psychosis, CR addresses cognitive impairments while CBTp improves psychopathology.

The key components of CBTp are engagement, collaboration and homework assignments to help the person become their own therapist and to be able to link behaviour with the thoughts and emotions. The therapy uses a problem-solving approach. The client is trained to think like a scientist looking for evidence, testing hypotheses and exploring the reality of their thoughts and perceptions.

The therapist uses a Socratic questioning approach to encourage the person to look for evidence, reason and rationale. Reality testing is used to actively find evidence to test the reality of a belief or an assumption. Furthermore, the therapist focuses on improving coping, building social and independent living skills to speed up recovery and an overall state of well-being. A behavioural activation approach is used to address negative symptoms. Thus, CBTp typically focuses on psychotic symptoms and functional outcomes. However, it does not address cognitive impairment in schizophrenia.

In spite of the robust evidence base and recommendations by the national organisations (e.g. National Institute for Health & Care Excellence [NICE] in the United Kingdom, the American Psychiatric Association [APA] and the Royal Australian and New Zealand College of Psychiatrists) availability of CBTp remains limited due to resources. There is currently an emphasis on providing low-intensity CBTp in brief, or even self-help and guided self-help formats. A recent meta-analysis reported moderate effect size for brief CBTp compared with treatment as usual and a smaller effect size for brief CBTp compared with other treatments (supportive counselling and befriending). None of the trials compared brief CBTp with the standard CBTp. The brief CBTp had a larger effect on negative symptoms compared with the positive symptoms. Most importantly, the effect was maintained on follow-up, which ranged from 6 to 18 months (Naeem et al., 2016b).

Other low-intensity approaches include guided self-help and self-help, especially cognitive behavioural therapy (CBT)-based self-help, delivered through mobile health platforms. We recently published a feasibility trial of cognitive behavioural–guided self-help for psychosis. The intervention had a positive effect on positive and negative symptoms, general psychopathology and measures of disability (Naeem et al., 2016a). However, individuals with a moderate degree of psychopathology and relatively low level of disability were recruited for this trial. This decision was based on our observation during the intervention development phase that this intervention is less likely to work with persons with schizophrenia who are cognitively impaired. Subsequent experience and feedback suggested that it might be useful for those with lower levels of cognitive impairment and disability.

Individuals with schizophrenia often experience a decline in neurocognitive abilities such as working memory, attention, planning, problem-solving and cognitive flexibility. This might lead to poor engagement with the CBTp. The aim of CR therapy, also referred to cognitive enhancement therapy or cognitive rehabilitation, is to improve the cognitive impairment. It is postulated that improvement in cognitive functions will enhance social functioning. This can potentially help prepare a larger number of individuals with schizophrenia for CBTp.

A variety of methods of CR is practised. Most of the CR programmes combine the rehearsal learning also called ‘drill and practice’ with ‘strategy coaching’ in which the participants are helped to develop strategies to undertake cognitive tasks. These CR strategies contribute to improving the underlying neuropsychological functions that help persons with schizophrenia think, concentrate and learn. However, CR does not directly address the psychopathology (e.g. delusions or hallucinations).

Recently, CR has been delivered using mobile health platforms, which has the advantages of being standardised and more efficient. This approach also allows therapy to be extended through independent exercise on a home computer. Computer-delivered programmes are acceptable to participants, especially younger people for whom these programmes resemble recreational computer games and can be delivered with minimum training.

There is now sufficient evidence that CR works. A meta-analysis (Wykes et al., 2011) reported that CR could have long-lasting effects on global cognition and functioning. However, the effect on the psychotic symptoms was small and disappeared at the follow-up assessment. There was no association between various elements of the treatment (e.g. remediation approach, duration, computer use) and cognitive outcome.

It can be seen that both CBTp and CR therapy are evidence-based approaches, and that each intervention targets different but complementary aspects of schizophrenia. While CBTp teaches persons with schizophrenia to challenge and cope with the psychotic symptoms, cognitive enhancing interventions aim to target (and subsequently augment) cognitive functions to support real-world function and might better prepare individuals to learn and practise the CBTp techniques. It, therefore, seems natural that the two therapies should be combined. Ideally, the CR is provided before CBTp. Drake et al. (2014) reported a trial of computerised CR compared with time-matched social contact (SC) preceding CBTp for persons with first-episode non-affective psychosis. Although there was no difference between the two groups in terms of psychotic symptoms, combined intervention was linked to improved insight. The most significant finding was that after CR, CBTp was shorter (median 7 sessions) compared with SC group (median 13 sessions). It is possible that reduction in number of the sessions was due to improvement in cognitive functions in those who received prior CR. Therefore, CR can be used prior to CBTp in brief and self-help formats to achieve rapid gains from therapy and to improve recovery.

We can summarise that, although limited availability of CBTp can be overcome by providing low-intensity and self-help CBTp, cognitive impairment can be a hindrance to these low-intensity formats. A course of CR before a low-intensity or self-help CBTp might improve outcomes. Therefore, combining CR and the low-intensity or self-help CBTp might offer the best way forward in providing CBTp in low-intensity and self-help formats.