Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

3.1 Introduction

Cognitive behavior therapy for psychosis (CBTp) was developed in the post-institution era. Despite the case studies by Beck (1952) and Shapiro and Ravenette (1959) in the 1950s, specific symptom interventions for schizophrenia did not appear until much later. The first controlled studies on cognitive behavior therapy for psychosis (CBTp) emerged in the early 1990s in the UK. It was felt that traditional therapies and antipsychotics at that time were not able to help patients with psychosis who needed more support with managing their distress, dealing with psychotic symptoms, and in improving their functioning in the community. Theoretical underpinnings such as the stress vulnerability models were developed to understand not only the development of the disorder but also its maintenance. These also began to be informed by research on expressed emotion (Butzlaff and Hooley 1998; Vaughn and Leff 1981) and so began to include social and psychological markers as well as biological ones. The difficulty in identifying rigorous and unambiguous psychosocial markers for psychosis may have hampered further development of this area initially, but the identification of ethnicity, deprivation, trauma, and drug misuse as key factors in its etiology and relapse has subsequently influenced interventions.

The 1990s was also the time when public pressure following some notorious scandals in the UK led to major changes in the health system, including the rise of clinical governance, need for evidence-based practice, and better coordination of care through the Care Programme Approach. Cognitive behavior therapy for anxiety and depression was recommended by the National Guideline development bodies (e.g., the UK National Institute for Health and Clinical Excellence, now called the National Institute for Health and Care Excellence), and during this time CBT also increased its theoretical research base. It was inevitable that eventually some of the developed techniques would be used for people with a diagnosis of schizophrenia. Since the emergence of the first controlled studies on cognitive behavior therapy for psychosis (CBTp) in the early 1990s in the UK, this treatment has developed further and included some of the theoretical underpinnings of CBT from other disorders. However, unlike CBT for other disorders which generally have its roots in the USA and particularly the Beck Institute in Philadelphia, CBTp developed predominantly in the UK (Wykes et al. 2008), and Beck himself was very supportive of it. Reviews of studies of CBTp have suggested that they are useful for the treatment of schizophrenia. Twelve to twenty sessions of CBT for psychosis (CBTp) is now recognized as an effective intervention for schizophrenia in clinical guidelines in the UK and the USA (Dixon et al. 2010; NICE 2014). There is evidence from meta-analyses, too, that it is effective. For example, Wykes and colleagues (2008) reported overall beneficial effects of the target symptom (33 studies; effect size 0.40 [95 % confidence interval) as well as significant effects on positive symptoms (32 studies), negative symptoms (23 studies), functioning (15 studies), mood (13 studies), and social anxiety (2 studies) with effect sizes ranging from 0.35 to 0.44.

3.2 What Happens in Cognitive Behavior Therapy for Psychosis

The cognitive theory proposes that through the examination of thought processes and by evaluating their accuracy, many negative emotional reactions due to inaccurate or distorted thinking could be reduced or extinguished (Beck and Emery 1979). The key elements of CBT for depression and anxiety include engaging the patient, collaboratively developing a problem list, and deciding on a clear goal for the therapy session (Beck et al. 1979). Homework assignments are agreed between the patient and the therapist. The aim of homework is to help the person become their own therapist and practically link behavior with thoughts and emotions. This essentially means that the patient uses self-help, and the therapist guides the patient during a session while the patient works as his own therapist during the period between two sessions. Regular feedback from the therapist and from the patient with regular summaries is also a crucial element. This therapy structure relies on collaborative working with the patient within an empirical methodology. The overall style of therapy is underpinned by a problem-solving approach that is grounded in reality. The patient is trained to think like a scientist looking for evidence, testing hypotheses, and exploring the reality of their thoughts and perceptions. A formulation is jointly generated to make sense of the emergence and maintenance of the problem at hand (Tai and Turkington 2009). The therapist uses guided discovery and Socratic questioning to identify distortions in the thinking style of the patient.

Communication in CBT uses a Socratic dialogue rather than therapist interpretation of thoughts, feelings, and behaviors. Beck et al. (1979) proposed that this method can be used to examine thoughts and beliefs for their accuracy. He demonstrated the usefulness of the Socratic questioning technique to encourage the probing of evidence, reason, and rationale. For example, a patient who believed that he was under surveillance would be asked to give a rationale for his belief. Other useful techniques include reality testing where a patient will be encouraged to actively find evidence to test the reality base of a belief or assumption. This is a process which is done in collaboration with the therapist, and behavioral experiments whereby a scientific experiment can be set up to test a specific prediction. For example, a person who believes that his next-door neighbor is constantly talking about him can be asked to record the conversation.

Cognitive behavior therapy for psychosis (CBTp) is based on these above principles. In particular, it heavily relies on the use of Socratic dialogue, guided discovery, reality testing, and behavioral experiments. It also uses work on improving coping (Tarrier and Calam 2002), building social and independent living skills, and increasing compliance using behavioral strategies such as linking medication taking to another activity. Similarly, negative symptoms were targeted by providing graded activity programs (Meichenbaum and Cameron 1973) and understanding them as protective and providing alternative strategies for them. These approaches have continued to be applied where deficit symptoms of schizophrenia and improving functional outcomes are the main focus of intervention. According to the cognitive model, hallucinations and delusions can occur when anomalous experiences that are common to the majority of the population are misattributed in a way that has extreme and threatening personal meaning (Garety et al. 2001; Morrison 2001). These models specify the role of faulty beliefs, increased attention to threat-related stimuli, biased information processing of confirmatory evidence, and safety behaviors (i.e., avoidance of specific situations) in the experience of positive symptoms. The emphasis is in the distress resulting not necessarily from difficult experiences, but the meaning placed on those very experiences. Cognitive theory is based on the notion that the cognitive processes implicated in mood and anxiety disorders occur transdiagnostically. Research findings support the notion that psychotic symptoms can be conceptualized with reference to normal psychological processes, whereby the content of symptoms is understandable and amenable to CBT (Allison Harvey 2004; Haddock and Slade 1996).

It was recognized that CBT for disorders such as anxiety and depression could be applied in schizophrenia with some key adjustments (Tai and Turkington 2009). Stigma was addressed by identifying the negative beliefs and assumptions people held about the diagnosis and prognosis of schizophrenia and then providing evidence that some of these experiences are actually fairly common in the general population, i.e., normalizing. In addition, the therapist provided alternative explanations, such as explaining the role of stress, that provided more optimistic and hopeful perspectives (Harrison et al. 2001) promoting beliefs in the potential for recovery. Compared with CBT for other disorders, the sessions were often shorter in length and much more flexible, and homework was simplified. The role of sleep disturbance, affect, and safety behaviors (e.g., behaviors such as avoidance that maintained faulty beliefs) was identified to produce mini-formulations of positive symptom maintenance (Morrison 2001). Cognitive biases are directly addressed by CBT predominantly through focusing on the content of thoughts and styles of thinking. These include the jumping to conclusions error and biases in styles of judgment found in individuals with unusual beliefs (delusions) and the biases in attributional styles and attentional processing associated with hallucinations. In CBT, it is the individual’s personal meaning, understanding, and coping with symptoms that are the focus of treatment. For example, individuals are facilitated in testing out the location of the hallucinations (internal vs external), carefully examining the appearance and behavior of suspected persecutors, and attempting homework that is pertinent to their stated goals (Tai and Turkington 2009).

3.3 Need for the Development of Brief CBTp

In spite of a strong evidence base, availability of CBTp remains an issue. This is especially true for the North America compared to the UK. It has been suggested that potential factors contributing to this difference include greater skepticism about the benefits of psychotherapy for persons with severe mental illness, overoptimism concerning the clinical benefits of polypharmacy, and the traditional separation between psychology and psychiatry in the USA as compared to the UK (Mueser and Noordsy 2005). A recently conducted survey of training directors in US psychiatry residency and clinical psychology doctoral programs to characterize the penetration of CBTp training and to assess their familiarity with basic CBTp facts reported that directors displayed limited knowledge of CBTp effectiveness, with only 50 % of psychiatry and 40 % of psychology directors believing that CBTp is efficacious. Only 10 % of psychiatry and 30 % of psychology directors were aware that the CBTp evidence base is based on meta-analyses (Kimhy et al. 2013). There are, however, limitations in availability of CBTp even in the UK, and the current evidence suggests that at most 50 % of those suffering from schizophrenia in the UK have access to CBTp (D. Kingdon and Kirschen 2006). This has led the UK government to introduce an “access and waiting standard” requiring services to offer CBT and family intervention within 2 weeks from referral for assessment of psychosis, beginning from April 2015.

Some approaches to increase the efficiency of CBT treatments include adapting individual treatments to a group format, self-help materials, bibliotherapy, and eMedia-assisted therapy programs. Other options might include training frontline staff in CBT interventions and utilizing families or peers in providing CBT interventions. The most common approach for enhancing efficiency, however, is to abbreviate existing CBT treatments by reducing the number of treatment sessions. There is evidence from research that brief CBT for psychosis can be delivered by community mental health nurses (Turkington et al. 2006). Brevity has many clear advantages. Increased cost-effectiveness could make treatment accessible to more individuals in need of assistance. Patients enjoy rapid treatment gains, and this may also improve the credibility of the treatment and increase the motivation for further change (Hazlett-Stevens and Dryden 2005). These efficient ways of delivering therapy have the potential to substantially reduce the gap if the effectiveness of a brief CBT can be demonstrated.

3.4 What Is Brief CBT for Psychosis

Currently there is more literature on brief CBT for depression and anxiety disorders than for schizophrenia. Standard CBT for depression is considered by most to be delivered between 10 and 20 sessions, but there is no agreement as to how many sessions should be included in brief CBT for depression (Bond and Dryden 2005). Churchill et al. (2001) described brief psychological interventions for depression to be delivered in 20 or fewer sessions, while Cully (2008) described brief CBT for depression to be delivered in between four and eight sessions. The Improving Access to Psychological Treatment (IAPT) services which exist across England offer a “low-intensity” range of treatments which are CBT based and meant for their use in the initial referrals with anxiety and depression; only if these are not successful, patients are offered standard CBT. In a recent Cochrane review, we have suggested that the standard CBT for psychosis (CBTp) involves around 16 sessions (12–20 sessions) over 4–6 months, while brief CBTp involves around 6–10 sessions in less than 4 months (Naeem et al. 2014a). This cutoff was based on the observation that current standard CBTp treatments typically span 12–20 sessions over 4–6 months (NICE 2014). We located empirical studies of the efficacy of brief CBTp by asking experts in a variety of areas about available research and by searches of Psychological Abstracts. In the absence of a clear definition for brief CBT for schizophrenia, we adopted our definition after a careful review of the literature on brief CBT for depression.

Although there is no difference in terms of theory or technical application of brief therapy compared with normal CBT, brief CBT focuses on more intensive work. Therefore, it might not be suitable for every patient with psychosis. There might also be differences in therapy when delivered by expert therapists compared with nonexperts. Brief CBT puts a greater burden on the patient to engage actively in treatment both during and between sessions. It can be argued that in psychotic disorders, especially with symptoms such as severe auditory hallucinations and high conviction delusions, brief therapy may not work as engaging patients and overcoming psychotic phenomenon may need a prolonged and intensive treatment approach. It is also possible that brief therapy may leave patients more confused and could prove harmful. However, the literature on the use of brief CBT for psychotic disorders does not seem to substantiate these apprehensions. There is some evidence to suggest that brief therapies are effective for clients with psychosis, but the research in this area needs to address the fundamental issue of the dose-effect relationship in CBT. We have described the three major trials of brief CBTp in the next section to give the reader a better understanding of these brief therapies.

3.5 Evidence from Research

In a recent review of the literature, we identified nine papers (covering seven studies) that included 1,207 participants, 636 in CBTp arm and 571 in the comparison arms. Brief CBTp showed moderate effect sizes (Hedge’s g = 0.43) (Naeem et al. in press). In this review we used our previously suggested criteria flexibly. We will briefly describe four studies, which fully fulfill the criteria for brief CBTp that we have proposed.

3.6 Description of Studies That Fulfill the Criteria for Brief CBTp

There are three randomized controlled studies of brief, six-session CBT currently published which are modularized but with different approaches. The Insight study (Malik et al. 2009; Rathod et al. 2005; Turkington et al. 2002, 2006) used the approaches described for CBT for psychosis but reduced the number of sessions to six and included three sessions with carers, where the client agreed to this. The other study reported a culturally adapted CBT for psychosis from Pakistan that delivered six sessions of therapy (Naeem et al. 2015). The Worry Intervention Trial (WIT) (Freeman et al. 2015) for paranoia involved an assessment of the current state and symptoms but then focused directly on worry and used techniques to address this specific issue. Each will be described and then proposals for a phased approach to the use of brief and standard CBT for psychosis described. The fourth study reported a group CBTp approach for hallucinations (Wykes et al. 2005).

3.7 Insight Study

The Insight study was a multicenter randomized controlled trial which compared a brief CBT intervention and carer sessions with treatment as usual. It had wide inclusion criteria: anyone with a diagnosis of schizophrenia, schizoaffective, or delusional disorders and relatively narrow exclusions: unable to communicate in English and substance dependence (use did not lead to exclusion). Four hundred twenty-two patients were recruited and 357 completed the study. The therapists were psychiatric nurses who were provided with 2–3 weeks therapy training and then weekly supervision. They were not experienced in using CBT and did not have CBT accreditation.

The results at 3 months (end of therapy) were a significant improvement in overall symptoms, insight, and depressive symptoms compared to treatment as usual. At 1 year, there were continuing positive effects on insight and depression and for negative symptoms. The effects on insight were for treatment and symptoms, but not into being ill at 3 months: the acceptance of the need for treatment was still present at 1 year. There were distinct effects on different cultural groups with those from black minority ethnic populations not doing as well which has led to a series of studies seeking to address this (Rathod et al. 2005). Further analysis established that those with mild-to-moderate drug misuse did no worse than others (Naeem et al. 2005) and that anxiety symptoms also improved (Naeem et al. 2006). Predictors of good outcome were higher levels of positive symptoms and insight (Naeem et al. 2008).

The therapy was adapted from that described in standard approaches involving 16–20 sessions. The initial phase involved assessment and engagement, followed by formulation using a “Making Sense” approach. This assembled information on current problems were identified with the client and then predisposing, precipitating, perpetuating, and protective factors (explained as what happened before and what happened when you became unwell, what’s kept it going, and what can stop it) were explored. Relevant connections between current and past issues and between the thoughts, feelings, and behavior were then made with consideration of relevant social circumstances and physical symptoms/illnesses.

Once these were established, the next phase was to work directly with the problems elicited using the formulation to make sense of them and then move to problem-solving. This usually involved direct work with voices, delusions, and negative symptoms. Work with voices would involve understanding and reattribution with normalizing to reduce fear and self-stigmatization. Coping strategies that were helpful were reinforced and further developed. Content of voices were discussed and debated such that, in some instances, clients could begin to enter into a constructive dialogue with their voices, reducing distress and working toward a correction of the power imbalance they experienced.

Delusions were explored with a focus on when the belief emerged allowing for a supported reassessment of the prevailing circumstances and events. Discussion included seeking relevant information about the focus of delusions, e.g., satellites or the Internet if these were considered to be causing concern. This process generally improved engagement and often led to behavioral changes, e.g., increased socialization. Changes in beliefs themselves were less likely within the short timescale of therapy but sometimes followed.

Work with negative symptoms accompanied approaches with other symptoms as simply equipping clients with an understanding of voices or paranoia, e.g., delusions of reference, and these interventions led to increased motivation and interaction with others. Specific long-term goals were discussed to instill hope and provide direction. Short-term goals were usually quite simple and readily achievable to build confidence and to instill a sense of self-efficacy.

Carer work was acceptable to many clients when an explanation of the reason for it was given – to help the carer understand and cope better – and agreement made about what should and should not be disclosed. It was usually agreed that the formulation could be shared – often with the client present – and then work on coping with specific symptoms and situations usually proceeded. Occasionally it involved negotiation between client and carer about specific issues but generally consisted of a sharing of information and problem-solving areas of uncertainty. Three sessions over the 3-month period of intervention was assessed as very valuable by carers and seen as sufficient by most. Clients were asked about how sufficient they found six sessions over 20 weeks, and while around 40 % were satisfied, a majority expressed that they would have liked more number of sessions.

3.8 Worry Intervention Trial (WIT) for Paranoia

The WIT trial randomized 150 patients in two centers to an assessment and then four sessions of a worry intervention (Freeman et al. 2015). The clients were included if they met criteria for paranoid delusions and worry. Very few patients referred with paranoia failed to reach the threshold for worry. Exclusion criteria included substance dependence. The therapists were clinical psychologists who had a brief training and weekly supervision in the use of the intervention.

The results were that the WIT group improved significantly in terms of worry, well-being, and delusional conviction compared to treatment as usual. The reduction in worry was a mediator of the effect on delusions.

The intervention is based on that used in generalized anxiety disorder which was found in a pilot study to be effective in paranoia. The theoretical basis for this had been well established by Daniel Freeman (Foster et al. 2010) but required substantive evaluation in a fully powered RCT. The therapy included six modules covering the following: assessment of the situation and symptoms, then focus on worry – whatever has happened, worrying about it is certainly not helping – let’s see if we can do something about that. Psychoeducation about worry was provided with review of positive and negative beliefs about worry. An increased awareness of the initiation of worry was developed with the identification of individual triggers. The client was encouraged to learn to “let go” of worry with the use of worry periods. Problem-solving was substituted for worry and clients were encouraged to use relaxation exercises. A simple individualized formulation of each person’s worry was developed and homework between sessions was agreed. Written information was provided in the form of leaflets about relevant issues, e.g., “winning against worry.”

The therapy was remarkably well accepted by patients, and recruitment to the study was straightforward with the recruitment target reached within the required timescale. Dropout rates were very low – less than 5 % – and results were similar in both centers, Oxford and Southampton (UK). It therefore seems to be an intervention which is generalizable and applicable in clinical practice.

3.9 Brief Culturally Adapted CBT for Psychosis

In this study brief culturally adapted CBT for psychosis (CaCBTp) (Naeem et al. 2014b, 2015) was adapted on the basis of qualitative work with service users, professionals, and carers. The intervention was targeted at symptoms of schizophrenia for outpatients from these groups compared to treatment as usual (TAU). A total of 116 participants with schizophrenia who were recruited from two hospitals in Karachi, Pakistan, and randomized into two groups with 1:1 allocation (CaCBTp plus TAU = 59, TAU = 57). A brief version of CaCBTp (six individual sessions with the involvement of main carer, plus one session for the family) was provided over 4 months. Psychopathology was measured using the Positive and Negative Syndrome Scale (PANSS) of schizophrenia, the Psychotic Symptom Rating Scales (PSYRATS), and the Schedule for Assessment of Insight (SAI) at baseline and end of therapy. Participants in treatment group showed statistically significant improvement in all measures of psychopathology at the end of the study compared with control group. Participants in treatment group showed statistically significant improvement in positive symptoms (PANSS, Positive Symptoms Subscale; p = 0.000), negative symptoms (PANSS, Negative Symptoms Subscale; p = 0.000), delusions (PSYRATS, Delusions Subscale; p = 0.000), hallucinations (PSYRATS, Hallucination Subscale; p = 0.000), and insight (SAI; p = 0.007). The results suggest that brief, culturally adapted CBT for psychosis can be an effective treatment when provided in combination with TAU, for patients with schizophrenia in a low- and middle-income country (LAMIC) setting. This is the first trial of CBT for psychosis from outside the Western world. These findings need replicating in other low- and middle-income countries.

Therapy was provided according to a manualized treatment protocol (Kingdon and Turkington 2002). Therapists were three psychology graduates with more than 5 years experience of working in mental health, who were trained by Farooq Naeem (FN). An important part of cultural adaptation of the CBT for psychosis is the involvement of the family member (Naeem 2013). Families are heavily involved in patient’s care and serve as the main caregivers to psychiatric patients in Pakistan, and, through our experience of adaptation of CBT for Pakistan, we understand that their involvement can enhance the acceptability of treatment. Therefore, this brief version consisted of six sessions for participant plus one session for the family. Every participant was accompanied by a carer who acted as co-therapist. Although therapy was provided flexibly, the sessions typically focused on the following:

  1. 1.

    Formulation and psychoeducation

  2. 2.

    Normalization and introduction to stress vulnerability model

  3. 3.

    Working with hallucinations

  4. 4.

    Working with delusions

  5. 5.

    Working with negative symptoms

  6. 6.

    Termination work and relapse prevention

Therapy was delivered using guidelines developed for cultural adaptation in Pakistan. These guidelines were developed in our preliminary work, in which CBTp was adapted using a series of qualitative studies similar to those we used for cultural adaptation of CBT for depression. During this preliminary work to adapt CaCBTp for use in Pakistan, we explored the views of patients, their carers, and the health professionals in this area. A total of 92 interviews was conducted by three psychologists. We conducted qualitative interviews with mental health professionals (n = 29) and patients (n = 33) and their carers (n = 30). The results of the mentioned studies highlighted the barriers in therapy (e.g., lack of awareness of therapy, family’s involvement, traveling distance and expenses, and uncooperative family caregivers) as well as strengths while working with this patient group. Patients and their carers in Pakistan use a bio-psycho-spirituo-social model of illness. They seek help from various sources, including faith healers. Therapists did make minor adjustments in therapy.

In addition to one additional session for the whole family and the involvement of a carer throughout the therapy, other salient cultural adaptations that we incorporated in the CBT manual were the following:

  • A spiritual dimension was included in the formulation, understanding, and in planning the therapy.

  • Urdu equivalents of CBT jargons were used in the therapy.

  • Culturally appropriate homework assignments were selected, and participants were encouraged to attend even if they were unable to complete their homework.

  • Folk stories and examples relevant to the religious beliefs of the local population were used to clarify issues.

3.10 Group CBT for Voices

Wykes et al. (2005) conducted a study of group CBT for voices that was delivered in seven sessions. The therapists who carried out this therapy were drawn from local services and then trained in group CBT techniques. Many but not all were experienced in providing individual CBT. Group CBT for the positive symptoms of psychosis provided the four key elements of CBT: engagement, collaborative discussion about an agreed model, cognitive restructuring of delusional beliefs, and reducing negative self-evaluation. Group CBT for voices used a manualized therapy for seven sessions each having a specific goal. The sessions included (i) engagement and sharing of information about the voices, (ii) exploring models of psychosis, (iii) exploring beliefs about hallucinations, (iv) developing effective coping strategies, (v) how to improve self-esteem, (vi) developing an overall model of coping with voices, and (vii) following up session.

Participants were included in this study, only if they had a diagnosis of schizophrenia and experienced distressing auditory hallucinations (rated on the PANSS). They were randomly allocated to group CBT (n = 45) or a control group who received treatment as usual (n = 40). The two main outcomes were social functioning as measured by the Social Behavior Schedule and the severity of hallucinations as measured by the total score on the Hallucinations Scale of PSYRATS. Assessments were carried out at baseline, 10 weeks (post therapy) and 36 weeks (6 months following therapy).

Results: Mixed random effects models revealed significant improvement in social functioning (effect size 0.63 six months after the end of therapy). There was no general effect of group CBTon the severity of hallucinations. However, there was a large cluster effect of therapy group on the severity of hallucinations, such that they were reduced in some but not all of the therapy groups. Improvement in hallucinations was associated with receiving therapy early in the trial and having very experienced therapists (extensive CBT training which included expert supervision for a series of individual cases for at least a year following initial training).

3.11 Summary

To summarize, there is now some evidence to suggest that brief CBT for psychosis can be effective in treating symptoms of schizophrenia. These interventions can form the basis for establishing an evidence-based treatment strategy for CBT for psychosis to assist in meeting implementation targets, e.g., the English “access and waiting standard” of treatment for psychosis within 14 days of presentation (https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/361648/mental-health-access.pdf). They can form the first phase of intervention using trained mental health practitioners (Insight) and clinical psychologists (WIT, culturally adapted CBT). Whether WIT can be delivered successfully by other trained practitioners has yet to be demonstrated, although proposals for such a study have been prepared for submission for funding. Delivery of CBTp using a stepped-care approach is in line with the current model of service delivery and is a major development in delivering CBTp for a very deserving population.

3. Conclusion

Brief CBT for psychosis is a developing area. Furthermore, research needs to be done to compare brief with standard CBTp. There is also a need to further explore the possibility of delivering CBTp using self-help, guided self-help, and eMedia. One important area in which brief interventions can be tried is case management, where frontline workers can be easily trained to deliver therapy. This has enormous implications for service delivery, reduction of distress and disability in clients with psychosis, and, above all, in terms of costs. Ideally CBTp should be provided in a variety of formats, with the dose of therapy being increased for those who do not respond to low-intensity approach.