 |
 |
|
|
|
 Cognitive-behavioural therapy for psychosis, which was designed to treat positive psychotic symptoms, has clear positive and lasting effects on delusions. With regard to its effectiveness on voice hearing, however, reported results have been equivocal (1–4). Therefore, HIT is designed to focus specifically on auditory vocal hallucinations. HIT has significant benefits for psychotic symptoms, subjective burden, quality of life, and social functioning (5–7) in TRVH patients. This study evaluated the cost-effectiveness of the HIT program for drug-refractory voice hearing schizophrenia patients over an 18-month follow-up period. Preliminary findings on patient outcome from the treatment phase of the study, as well as the 18-month outcomes on cost-effectiveness and quality of life, were previously reported (8–10). The current study aimed to investigate whether HIT had a beneficial effect on psychopathology, subjective burden, and control of voice-hearing outcomes over a period of 18 months. MethodThis RCT was carried out in 4 centres in 3 Dutch provinces with patients allocated to either HIT or RT groups. Outcome data were collected at baseline (pretreatment), 9 months (posttreatment), and 18 month (follow-up). Subjects The medical ethical board of the University Hospital Groningen approved recruitment and study procedures. Inclusion criteria were as follows: persistent auditory hallucinations for over 2 years despite adequate treatment; a DSM-IV diagnosis of schizophrenia, schizoaffective, or psychotic disorder NOS; former use of at least 2 AP drugs in adequate dosages for an adequate period of time according to the DPA guidelines; no previous CBTp treatment for auditory hallucinations; no current abuse of psychoactive drugs or alcohol (some use of cannabis and alcohol was allowed); and an estimated IQ of above 80. Mental health care institutions and case managers of community care teams screened their caseloads for eligible patients, informed them of the study, and asked their permission to be contacted by a researcher. The researcher subsequently contacted patients and gave them full information about the study and the consequences of participation. One hundred patients were approached for a SCAN interview (11). Twenty-four were excluded owing to (primary) substance abuse, absence of hallucinations, medication inconsistencies, wrong diagnosis, low intelligence, or incoherence. All eligible patients gave their written informed consent. An independent medical technology unit randomized patients. A minimization procedure, (12) that accounted for sex, duration of illness, severity of symptoms, and treatment centre, was applied. For reasons of statistical power, we aimed to obtain 80 patients. Ultimately, 76 were included (37 in the experimental condition). Subjects were diagnosed as having paranoid schizophrenia (78%), schizoaffective disorder (15%), or psychosis NOS (7%). Mean duration of hallucinations at baseline was almost 12 years, SD 10.4. Lifetime average number of hospital admissions was 3. The mean age of all patients was 36 years, SD 11.2. Of patients, 54% were men, 42% lived alone, and 33% lived with others in the community (parents, partner, or others), while 24% lived in sheltered accommodations. Only 3 patients (4%) had a paying job, while 84% received welfare, unemployment assistance, or a disabilities pension. The experimental and control groups did not differ significantly on any of the relevant baseline variables. Interventions HIT integrates cognitive-behavioural interventions, coping training, FT, rehabilitative efforts, mobile assertive crisis intervention, and AP medication with RT. Both patients and caregivers receive cognitive interventions and coping training. The use of attitudinal and motivational techniques, for example, selective motivational interviewing, paradoxical interventions, an attitude of acceptance and negotiation, positive labelling of symptoms and behaviour for enhancing compliance, and the “two-realities-principle” that accepts voices as factual experiences and relies on patient expertise and responsibility, are important additional features of HIT. These features help strengthen the potential of the cognitive and coping interventions. The cognitive interventions focus on triggering events, cognition, emotional and behavioural (re)actions, and consequent reactions of others. Patients are encouraged to develop coping strategies. They are also encouraged to monitor the voices and triggering events, the intensity of experienced interference, the effectiveness of actual coping behaviour, and the reactions of others. Relatives are invited to attend all meetings and to take an active part in the treatment process. They monitor their interactions with the patient and are trained in effective coping behaviour and selective reinforcement. The median duration of HIT is approximately 9 months, with a median of 11 contacts. Variation in duration (SD 9) and contacts (SD 12) can be large if need persists (9,10). All sessions are conducted as singlefamily intervention. Treatment focuses on “regaining control in my head.” HIT is given according to a protocol delineated in a manual but is tailored to individuals, which means that specific elements may vary in duration or intensity (13). No attempt was made to standardize RT. In The Netherlands, RT is mainly provided by regional multidisciplinary teams that make home visits, and provide psychiatric, social, and financial management, maintenance of medication, and ambulatory crisis intervention 24 hours daily. Optional interventions include supporting relatives, activities concerning housing, involvement in society, and rehabilitation. For optimal comparability with the experimental condition, RT therapists were asked to involve patients’ families. RT was continued in control patients. Experimental patients were allocated to HIT therapists (psychiatrists and psychologists) for 9 months. They were advised to continue practising their coping strategies and were then referred to their original RT therapists. Practitioners working with each treatment group were equally qualified in terms of Dutch board certifications, special courses, and years of experience treating schizophrenia patients. All were experienced in both hospital and community psychiatry. Medication was prescribed according to the guidelines of the DPA. Therapists working with the experimental group took a 5-day course in HIT, followed by biweekly live supervision for approximately 5 months prior to the RCT. Therapists provided either RT or HIT. Treatment fidelity in both conditions was satisfactory. Psychiatrists of RT patients monitored treatment fidelity in weekly case reviews. In the HIT program, adherence to the protocol was controlled in a systematic manner (14). Sessions 1 to 3 for the HIT patients were observed through a one-way mirror in addition to weekly case reviews and biweekly group supervision with supplementary consultations when therapists faced difficulties in adhering to the protocol. Outcome Measures The assessments were conducted by 2 independent assessors. They received formal training in the use of these instruments and have extensive experience with them in schizophrenia research. Their interrater reliability scores during regularly held in-house sessions made additional testing for this study superfluous. Patients primarily experienced changes in voice hearing and related subjective burden. Psychopathology was measured with the PANSS, a 30-item, semistructured interview on psychiatric symptoms over the 2 preceding weeks that measures psychopathology on a 7-point scale (15). The analyses used 5 dimensions: positive and negative symptoms, disorganisaztion, depression, and excitement (16). Total and general psychopathology scores were used as measures of mental health. Burden was measured with the AHRS, a 12-item questionnaire, administered as an interview, measuring characteristics of voices on a 5-point scale (17). Secondary outcomes were also used, including use and dosage of classical and atypical APs, benzodiazepines, and antidepressants. The amount of AP was converted into haloperidol equivalents. Adherence was recorded in terms of failures both to consent to treatment and to comply with the study. Analyses Principal component analysis with Varimax rotation and Kaiser normalization was applied to the AHRS data before treatment. Four indices emerged, including frequency and duration, amount and threat of negative content, distress (amount and intensity of distress and interference with daily functioning), and control. The first 2 indices were calculated by means of multiplying and the latter by summing the separate items. ITTs were performed on all 63 patients who completed assessments. We performed analyses with SPSS 12.0 for Windows, using the general linear model with repeated measures on the 3 assessments and testing the differential effects of treatment condition with alpha set at 0.05 (2-tailed). Sex, age, education, and illness (hallucination) duration were used as covariates. The correlation between medication and psychopathology score was calculated with Pearson’s rho. ResultsAt the time of the 18-month follow-up, 5 patients refused to participate, in addition to the 8 who dropped out during the treatment phase. Patients who dropped out did not differ significantly on any of the baseline variables. Eventually, 63 patients (83%) completed all 3 waves. Analyses were based on these patients. Patient outcomes Subjective Burden. At 18-month follow-up, both groups appeared to have improved in each domain of subjective burden (Table 1). Improvements were greatest in the experimental group, particularly with respect to negative content, distress, and total burden.
Psychopathology. Table 2 gives the PANSS scores for both groups. Groups differed in their progression of change. The patients in the control group improved during follow-up but not during treatment. The experimental group showed substantial improvements during the treatment phase that continued during follow-up. Improvement in favour of the HIT condition was significant at the 5% level in positive symptoms and at the 10% level in depression, general psychopathology, and total score. Differences on the disorganization factor only relate to the posttreatment period.
Medication. At baseline and posttreatment, differences in mean dosages of APs, antidepressants, and benzodiazepines were not statistically significant. At 18 months, no significant difference between groups was found in percentage of patients taking benzodiazepines (c2 = 1.06, df 1, P = 0.30) or antidepressants (c2 = 0.03, df 1; P = 0.88). However, mean dosages of AP medication were significantly higher in the control group (t = –2.70; df = 56; P = 0.009), while difference in the ratio of prescribed classical to atypical APs was not statistically significant (c2= 0.13; df = 2; P = 0.94). Neither the Pearson correlation of mean dosages of APs, nor any of the 5 PANSS factors appeared to be of great significance. DiscussionIntegrative treatment focusing on TRH is quite novel. We had a four-fold rationale for developing HIT. First, innovative treatment is needed because most patients with schizophrenia have disabling residual symptoms and are handicapped in social functioning, despite recent therapeutic advances (27). Second, optimal treatment for this population requires integrative treatment programs (28). Third, the alarming number of schizophrenia patients who reject treatment and the high rate of noncompliance (19–21,29–31) require specific strategies, such as motivational interviewing, positive labelling, paradoxical interventions, and the “two-realities” approach. Finally, although its effectiveness with regard to hallucinations and social functioning is equivocal (1,3,4,18), as it is the preferred treatment for delusions, one may expect positive results from CBTp (1,2,4,19,20,29) on delusional explanations of hallucinated voices. The comprehensiveness and duration of our program, its flexibility and individualized tailoring, the attitude and the single-FT modules, and the “selling” of rehabilitation as a form of coping differentiate HIT from most voice-focused CBTp programs (1–3,18). The number of sessions are equal in most CBTp and HIT studies. Duration of treatment is longer in HIT (9 months), while the rate of attrition is higher in most CBTp studies (30). This study included chronic psychiatric patients suffering from drug-refractory psychotic voices. At baseline, there were no statistical or clinical differences between the groups for any of the assessed variables, with the exception of self-care. Our patient population is comparable to most CBT studies for schizophrenia (1,2,18,19,25,26) but differs from the CBT studies (20,21) that included patients with acute and first-episode or affective psychosis.
Implementation of HIT in community-based facilities and low attrition may be considered strengths of this study. Cost effectiveness (8) and good consumer satisfaction are additional strengths. The attrition rate in our study is relatively low: 12 patients (16%) refused assessment at 18-month follow-up. Although the assessors were not involved in treatment, they could not be kept blind to it, which is a serious limitation. Contrary to instructions, patients unmasked their treatment condition. We recommend an additional study under blinded conditions, but this may prove an idle wish. Other limitations concern the exclusion of primary comorbid severe substance abuse and low IQ, which may have eliminated uncooperative patients. After treatment, improvements in subjective burden (distress and negative content) and psychopathology (positive symptoms, general psychopathology, and total PANSS score) were statistically significant in favour of integrative treatment. At 18 months, experimental patients retained improvements in subjective burden and positive symptoms at the 5% significance level, while significance of improvements changed to the 10% level in general psychopathology and total PANSS score, owing to improvements in the control group during the follow-up period. It appeared that, for approximately 14 months, 9 control patients received RTplus that mimicked both CBT and HIT but was lacking a formal treatment protocol. All 9 received additional coping training and symptom-oriented psychoeducation. Seven of them received extra CBT and 3 of them, caregiver interventions. Selection of these 9 was not random but depended on the motivation of both patient and therapist. RTplus patients improved significantly more than other control subjects. Within-group analysis of the RTplus patients indicated statistically significant improvements on positive symptoms (t = 2.36; P < 0.05) and depression (t = 3.48; P < 0.01) at 9 months, and on general psychopathology (t = 3.44; P < 0.05) and total score (t = 2.87; P < 0.05) at 18 months. Over the total 18-month period, there were substantial improvements on positive symptoms (t = 2.57; P < 0.05), depression (t = 4.16; P < 0.01), general psychopathology (t = 3.44; P < 0.01), and PANSS total score (t = 2.87; P < 0.05). One may argue that the effectiveness of RTplus points toward a lack of specificity of HIT. However, one may also argue the opposite, because the enrichments consisted of one or more HIT modules. Nevertheless, further research about HIT specificity seems warranted. Assuming that the RTplus group is not representative of RT, additional analyses compared HIT patients with remaining control subjects (n = 54, excluding the RTplus patients). These comparisons showed significant improvements in favour of the experimental group in control of voices, positive symptoms, depression, general psychopathology, and PANSS total score. Experimental and control conditions in psychosocial intervention studies in schizophrenia, including our own, vary in content, comprehensiveness, intensity, and duration, which may limit the comparability of findings (22,23). This raises questions about the relative impact of design, sequence, integration, and duration of psychosocial modules. Our study did not focus on these issues; therefore, we can only speculate on the relative effectiveness of various treatment components. Medication At baseline and posttreatment, differences between groups in mean dosages of benzodiazepines, antidepressants, and APs were not significant. By 18 months, the mean dosages of APs differed significantly in the treatment groups: prescription of APs increased in the control group, compared with a decrease in the experimental group. Groups did not significantly differ in the proportion of prescribed atypical APs, and mean dosages of AP were not significantly related to PANSS scores. Assuming those groups did not significantly differ in actual use and compliance, significant improvements in favour of HIT occurred independent of medication. Psychosocial Interventions, Subjective Burden, and Psychopathology From the above, one may infer that the effect of psychosocial interventions is crucial. The sustained improvement in voice hearing and social functioning in our experimental group suggests a more robust and generalized effect than that found in CBTp studies; for example, Haddock and others (1998) conclude “the results . . . confirm previous observations of the difficulty of treating hallucinations with CBT”(1). In another study, the modest benefits of CBTp over supportive counselling on voice hearing observed at 9 months lost significance at 12 months (4). After CRT, reduction of voices did not significantly differ from RT, either posttreatment or on follow-up (18), while “no changes in social functioning were directly associated with it” (3). Comparing CBTp studies with this study demonstrates differences in effect size in favour of HIT on positive symptoms and on social functioning after treatment as well as at follow-up (1,2,4,7,9,10,19). We could not satisfactorily explain why the difference between treatment conditions in ITTs (n = 63) at 18-month follow-up remained significant for subjective burden, quality of life, and social functioning but not for other aspects of psychopathology. HIT and RTplus patients improved equally, but assignment to RTplus lacked randomization, and its exposure time was 50% longer, warranting further investigation of its effectiveness. Further investigation is especially warranted because the effectiveness of both conditions was assessed after 18 months but at different moments in the treatment process, that is, after the treatment in RTplus and on follow-up in HIT. Conditions also differed in continuity of treatment and therapist. Subjective burden, control of voices, and psychopathology improved after both RTplus and HIT, suggesting that their shared modules—CBTp, FT, and coping training—either alone or in combination, are of the utmost importance to patient improvement. However, a single CBTp effect is unlikely in view of the already-mentioned CBTp studies. The effectiveness of FT psychoeducation plus coping training was demonstrated in schizophrenia patients (24). We hypothesize that the role of coping training in our program (that is anxiety reduction plus focusing plus distraction),which was appreciated and experienced as very helpful by HIT and RTplus patients as well as their relatives, is pivotal. Different from voice-oriented CBTp programs (2,25,26), HIT resulted in an enduring reduced PANSS depression score. Bentall and others (1994) found that distraction and focusing are equally effective coping strategies in reducing positive psychotic symptoms. Distraction, however, only gave depression as a side effect. Social Functioning and Quality of Llife Improvements in social functioning in most studies, if any, are limited. The effectiveness of CBTp on the social functioning of voice hearers appears equivocal. HIT induced enduring improvements in quality of life and social functioning (10). At Month 18, a 16% relative risk difference for persistent serious social disabilities was calculated with a number-needed-to-treat of 6. In conjunction with the rehabilitation module, FT may be considered responsible for good treatment compliance and improved quality of life and social role functioning. Active involvement of relatives in HIT is highly valued by both patients and family. It is unlikely that positive expectations of HIT have influenced results because HIT was hardly known at the time and was certainly not considered superior by patients or therapists. Thus far, we have noticed that HIT is also feasible and effective for inpatients, provided that attitudinal aspects of HIT can be implemented on the ward. The issues of dosage specificity and relative effectiveness of the various treatment components remains a subject of further research. ConclusionComparing the HIT program with RT demonstrates significant posttreatment improvements in favour of HIT in subjective burden, psychopathology, quality of life, and social functioning. Over time, improvements in these domains remained at the 5% significance level. Satisfaction of patients and relatives with, and adherence to, HIT has been fairly good. HIT appeared to be cost-neutral. Funding and SupportThe research is supported by an unrestricted grant from the Ziekenfondsraad in Amstelveen (OG98-040) and approved by the Medical Ethical Board of the University Hospital Groningen (MEC 97/02/029). References1. Haddock G, Tarrier N, Spaulding WD, Yusupoff L, Kinney C, McCarthy E. Individual cognitive-behavioral therapy in the treatment of hallucinations and delusions: a review. Clin Psychl Rev 1998;18:821–38. 2. Chadwick P, Lees S, Birchwood M. The revised beliefs about voices questionnaire (BAVQ-R). Br J Psychiatry 2000;177:229–32. 3. Wykes T, Reeder C, Williams C, Corner J, Rice C, Everitt B. Are the effects of cognitive remediation therapy (CRT) durable? Results from an exploratory trial in schizophrenia. Schizophr Res 2003;61:163–74. 4. Valmaggia L, Gaag van der M, Tarrier N, Pijnenborg M, Slooff C. Cognitive-behavioural therapy for refractory psychoatic symptoms of schizophrenia resistant to atypical antipsychotic medication. BJP 2005;186:324–30. 5. Jenner JA, van de Willige G, Wiersma D. Effectiveness of cognitive therapy with coping training for persistent auditory hallucinations: a retrospective study of attenders of a psychiatric out-patient department. Acta Psychiatr Scand 1998;98:384–9. 6. Jenner JA, van de Willige G. HIT: hallucination focused integrative treatment as early intervention in psychotic adolescents with auditory hallucinations. Acta Psychiatr Scand 2001;103:148–52. 7. Wiersma D, Jenner JA, van de Willige G, Spakman M, Nienhuis FJ. Cognitive behaviour therapy with coping training for persistent auditory hallucinations in schizophrenia: a naturalistic follow-up study of the durability of effects. Acta Psychiatr Scand 2001;103:393–9. 8. Stant AD, TenVergert EM, Groen H, Jenner JA, Nienhuis FJ, van de Willige G, and others. Cost-effectiveness of the HIT programme in patients with schizophrenia and persistent auditory hallucinations. Acta Psychiatr Scand 2003;107:361–8. 9. Jenner JA, Nienhuis FJ, Wiersma D, van de Willige G. Hallucination focused Integrative Treatment improves burden, control, and symptoms in schizophrenia patients with drug-resistant hallucinations, a RCT. Schizophr Bull 2004;30:127–39. 10. Wiersma D, Jenner JA, Nienhuis FJ, van de Willige G. Hallucination focused integrative treatment improves quality of life in schizophrenia patients. Acta Psychiatr Scand 2004;109:194–201. 11. World Health Organization. WHO SCAN: Schedules for clinical assessment in neuropsychiatry. Geneva: World Health Organization; 1992. 12. Altman DG. Practical statistics for medical research. London: Chapman and Hall; 1995. 13. Jenner JA, Gorkum Iv, Wiersma D. Coping training en cognitieve therapie bij schizofrenie patienten met chronische en invaliderende hallucinaties. Protocol van behandeling. Groningen: Afd Psychiatrie AZG/RuG; 1996. 14. Jenner JA. HIT: an integrative treatment for patients with persistent auditory hallucinations. Psychiatr Serv 2002;53:897–8. 15. Kay SR, Opler LA, Fiszbein A. Positive and Negative syndrome scale (PANSS) rating manual. San Rafael, California: Social and Behavioural Sciences Document; 1987. 16. Lindenmayer JP, Grochowski S, Hyman RB. Five factor model of schizophrenia: replication across samples. Schizophr Res 1995;14:229–34. 17. Haddock G, McCarron J, Tarrier N, Faragher EB. Scales to measure dimensions of hallucinations and delusions: The psychotic symptom rating scale (PSYRATS). Psychol Med 1999;29:879–89. 18. Wykes T, Thompson N, Hayward P. Changing voices: a randomized control trial of group cognitive treatment. Schizophr Res 2002;53:3–12. 19. Kuipers E, Garety PA, Fowler D, Dunn G, Bebbington P, Freeman D, and others. London-East Anglia randomized controlled trial of cognitive-behavioural therapy for psychosis. I: effects of the treatment phase. Br J Psychiatry 1997;171:319–27. 20. Drury V, Birchwood M, Cochrane R, Macmillan F. Cognitive therapy and recovery from acute psychosis: a controlled trial. Impact on psychotic symptoms. Br J Psychiatry 1996;169:593–601. 21. Kemp R, Hayward P, Applewhaite G, Everitt B, David A. Compliance therapy in psychotic patients: randomized controlled trial. BMJ 1996;312:345–9. 22. Dickerson FB. Cognitive behavioural psychotherapy for schizophrenia: a review of recent empirical studies. Schizophr Res 2000;43:71–90. 23. Gould RA, Mueser KT, Bolton E, Mays V, Goff D. Cognitive therapy for psychosis in schizophrenia: an effect size analysis. Schizophr Res 2001;48:335–42. 24. McFarlane WR, Dixon L, Lukens E, Lucksted A. Family psychoeducation and schizophrenia: a review of the literature. J Marital Fam Ther 2003;29:223–45. 25. Bentall RP, Haddock G, Slade PD. Cognitive behaviour therapy for persistent auditory hallucinations: From theory to therapy. Behav Ther 1994;25:51–66. 26. Haddock G, Morrison AP, Hopkins R, Lewis S, Tarrier N. Individual cognitive-behavioural interventions in early psychosis. Br J Psychiatry Suppl 1998;172:101–6. 27. Bustillo J, Lauriello J, Horan W, Keith S. The psychosocial treatment of schizophrenia: An update. Amer J Psychiatry 2001;158:163–75. 28. Brenner HD, Hirsbrunner A, Heimberg A. Integrated psychological therapy program: Training in cognitive and social skills for schizophrenic patients. In: Corrigan PW, Yudovsky SC, editors. Cognitive Rehabilitation for Neuropsychiatric Disorders. Washington (DC): American Psychiatric Press; 1996. p 329–48. 29. Pilling S, Bebbington P, Kuipers E, Garety P, Geddes J, Orbach G, Morgan C. Psychological treatment in schizophrenia: I Meta-analyses of family intervention and cognitive behaviour therapy. Psychological Medicine 2002;32:763–82. 30. Jenner JA, Wiersma D, Willige van de G, and Nienhuis FJ. Effectiveness of CBT for positive psychotic symptoms: Discrepancies between British and Dutch programmes. Schizophr Research 1999;36(1-3):326–7. 31. Young JL, Zohanna HV, Shepler L. Medication noncompliance in schizophrenia: Codification and update. Bull. Of the Amer Acad of Psychiatry and the Law 1986;14:105–12. Author(s)Manuscript received February 2005, revised, and accepted August 2005. 1. Associate Professor, Department of Psychiatry, University Medical Centre and Mental Health Care Foundation, Groningen, The Netherlands 2. Senior Researcher, Department of Psychiatry, University Medical Centre and Mental Health Care Foundation, Groningen, The Netherlands 3. Professor, Department of Psychiatry, University Medical Centre, Groningen, The Netherlands Address for correspondence: Dr JA Jenner, Department of Psychiatry, University Medical Center Groningen, PO Box 30.001, 9700 RB Groningen, The Netherlands e-mail: j.a.jenner@psy.umcg.nl
1 | 2
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||